By The Bell

The ABCs of VDI: user perception = reality

In the late 1990s, my company at the time focused on designing and deploying server-based computing (SBC) solutions using Citrix MetaFrame. One of our earlier implementations, at a small school district, included a classroom pilot with 25 Wyse WinTerms as well as 5 legacy PCs kept on hand to run the applications incompatible with Terminal Services. The teacher used one of these PCs, and her keyboard happened to break during the pilot. We plugged in a new keyboard which worked fine, and then plugged her old keyboard into a different unit to show that it was indeed broken. But she went around to all of the other teachers telling them, "Don't install Citrix in your classroom, it breaks keyboards." The project died.

Challenges with VDI

VDI utilizes a very different technology than SBC, but both host desktops on a central server farm controlled by the IT staff. Users become wary as they see the "personal" part of their PCs diminish. They tend to not like change – even when it's positive, and organizations that implement a VDI environment without careful consideration of how it will impact user perceptions do so at their peril.

Virtualizing desktops would be much easier if they could all be persistent. A persistent virtual desktop retains the user's profile and My Documents on a separate user data disk, enabling a very close approximation of the physical version. Maintaining separate operating systems and applications for each user, however, consumes expensive data center storage resources. Administrative operations such as patching and upgrades need to take place for each persistent image. The tie between the persistent profile and the underlying operating system makes tasks such as migrating users from Windows XP to Windows 7 very difficult.

The combination of inefficient resource utilization and management complexity can quickly negate the economic advantages otherwise enabled by a VDI environment, making the persistent virtual desktop model impractical for most situations. Organizations instead typically utilize non-persistent virtual desktops. Users, upon log-in, are assigned a virtual machine from a resource pool that utilizes Windows Roaming Profiles to establish personalized settings. But Roaming Profiles brings its own set of problems resulting from attempts to update user profiles from different sessions. And as user profiles grow, download times increase, network traffic becomes congested and profile corruption becomes much more likely.

VDI manufacturers have been furiously working to resolve the personalization problem. Citrix, for example, purchased a company called Sepago and now integrates the technology as Citrix Profile Manager. VMware announced its licensing of RTO's Profile Manager last year which will be integrated into an upcoming version of View. LiquidWare Labs recently added ProfileUnity to its portfolio. Newcomers Atlantis and Unidesk approach the personalization challenge using containers. AppSense originated the profile management category over ten years ago and continues to lead the industry with innovations such as an ability to manage hybrid environments consisting of virtual, physical and RDS (Terminal Services) architectures.

The Exploding VDI Industry

VDI started as a grass roots initiative around 2005 when VMware noticed its customers configuring desktop operating systems and applications as virtual machines on their ESX Hosts. Someone at VMware recognized the potential huge opportunity and coined the term, Virtual Desktop Infrastructure. An industry was born.

The increasing familiarity of IT staffs with server virtualization frequently emboldens them with the confidence to deploy a VDI solution; applying to the desktop the lessons learned from their server virtualization efforts. This approach inevitably leads to trouble when expanded to an enterprise initiative. Wildly varying usage patterns, for example, can quickly oversubscribe memory and disk IOPs – creating widespread havoc among users. Anti-virus scanning configured for a traditional physical environment can quickly bring VDI performance to its knees.

Gartner's March 2009 press release predicts exponentially increasing VDI sales of $65B by 2013, up from only $1.4B last year. The number of VDI disasters will skyrocket as well.

A Strategic Approach to Desktop Virtualization

Desktop virtualization demands a strategic approach incorporating the prerequisite assessment, planning, pilots and managed execution required for a successful enterprise project. Objectives, including projected costs and savings, should be identified prior to starting the design. The planning and proof-of-concept phases can then be initiated within the context of providing a virtual desktop architecture optimized to meet the organization's objectives. The architecture requirements largely determine the VDI components such as hypervisor platform, application virtualization, client devices, storage architecture, back-end virtualization infrastructure and many, many other considerations.

A key objective for a VDI implementation should be to provide a very positive user experience from the beginning pilot up through the production roll-out. Carefully planned deployments can generate a buzz among users, leading to an enthusiastic acceptance of virtual desktops. The appropriate personalization technology is one of the important elements enabling this success.

 

 

Microsoft licensing for VDI made really simple

The Microsoft licensing required for VDI is called VECD (Virtual Enterprise Centralized Desktop – formerly Vista Enterprise Centralized Desktop) and it replaces the need for either Windows 7 or Terminal Server (RDS). It is licensed only by physical end-point device and costs (list) $23 per PC or laptop per year if the organization has Software Assurance for the device (VECD for SA), or $110 per thin-client or zero-client device per year and per PC or laptop if not covered by Software Assurance (just VECD).

Additionally:

• Since VECD is licensed per device, multiple employees can use the same device without incurring extra licensing cost. This works out great for a training lab or for a factory floor where multiple users access a smaller number of devices.
• VECD includes home right use for primary users.
• A VECD license for a contractor accessing the VDI environment with her own device also applies to a replacement contractor as long as each only brings in one unique device.

Don’t believe any statistics you don’t make up yourself

 

"40% of all companies that experience a major disaster will go out of business if they cannot gain access to their data within 24 hours." 

Gartner (as reported on Rackspace's Web site)

This disconcerting statistic is widely referenced in promotional materials and presentations not only by Rackspace, but also by Double-Take, EMC, VMware, Frontline Data Services and many others. I've quoted it frequently in my own presentations over the years to emphasize the importance of utilizing virtualization to enable much faster data center recovery than is possible in the physical world. The only problem is that Gartner never said it.

The first reference I can find to the actual source is in a 2006 Citrix white paper, Avian Flu and the Opportunity to Get Ready, which attributed it to a survey by Eagle Rock Alliance. But the survey, 2001 Cost of Downtime, never reported the "40%" statistic. It simply asked the question of its 163 respondents, "At what point is the survival of your company at risk?" Thirty-nine percent reported a time-frame of 24 hours or less. The white paper apparently rounded this up to 40% and translated it into the percentage of companies that go out of business if they lose access to their data for over 24 hours. The statistic was then attributed to Gartner by someone else and now has become  disaster recovery folklore.

 

Thanks to Rick Tanner of LabCorp who diligently researched the "Gartner" statistic after attending my presentation at the CIO Forum in Raleigh last month.

VMware and Citrix both seek to upgrade XenApp customers to VDI

VMware and Citrix are both eager to upgrade Citrix XenApp customers to their respective VDI solutions. Former Citrix distributor, Alternative Technology Group of Arrow ECS (AltTech), with support from VMware, announced this morning its "Virtual Desktop Trade-in Program." As with Citrix's Trade-Up program, the Trade-In offer expires June 30, 2010. It offers XenApp customers a free upgrade to VMware View 4 Premier, but with an innovative twist that converts a normal CapEx budget expense into OpEx. Customers just make 36 monthly payments equivalent to the the Subscription Advantage costs they had been paying to Citrix.

Upgrading to View 4

The AltTech program originated with President & COO Bill Botti. AltTech was a prominent and specialized Citrix distributor for 13 years until Citrix terminated its relationship with all distributors other than Ingram Micro in 2008 (A rough chronology of events leading to the program is listed below). Botti was looking for a way to accelerate the growth of VMware View using existing Citrix Subscription Advantage expense each year to fund the project. In addition to aggressive pricing from VMware and aggressive financing from Arrow ECS Leasing Services, Botti wanted to make the migration planning easy. AltTech worked with Lakeside Software to use its SysTrack VMP (Virtual Machine Planner) to enable the analysis necessary to develop a clear migration plan.

AltTech estimates that XenApp customers spend on average around $75 per concurrent user per year in Subscription Advantage licensing to remain current. XenApp customers can instead spend the equivalent in monthly payments, without any cost for financing, over a three-year period and will receive a free VMware View 4 Premier license (also licensed by concurrent user). View 4 Premier includes the back-end vSphere Enterprise Plus licensing as well as ThinApp for application virtualization. The program, which requires customer credit approval, includes 3 years of basic Subscription and Support on the View 4 Premier licenses.

In order to assist with the conversion, customers will receive a free physical PC infrastructure assessment with SysTrack VMP from Lakeside Software. SysTrack VMP provides an integrated tool suite that provides assessment, capacity planning, virtualization modeling, predictive analysis, migration planning, storage throughput, space planning, power planning, firewall and latency analysis, user and application behavior analysis, pooled desktop / image planning, and application virtualization compatibility analysis.

Citrix Trade-up Program

Citrix's own recently announced trade-up programs seeks to migrate customers from XenApp to XenDesktop 4. Customers who upgrade all of their XenApp licenses at once receive either two XenDesktop user or device based license for every one XenApp concurrent license they trade-up. The trade-up includes 12 months of Subscription Advantage on XenDesktop 4.

Citrix's declining XenApp licensing revenues and increasing Subscription Advantage revenues give it a particularly strong interest in persuading customers to migrate to its VDI solution. But while XenDesktop tends to get great reviews (see InfoWorld's 2010 Technology of the Year Awards), analysts say that it runs a distant second to VMware in market share. VMware's dominance in the data center gives it an advantage with organizations wishing to run both their server and desktop virtual machines with the same platform and management console.

Zoaring Popularity of VDI 

The Enterprise and Platinum editions of XenDesktop now include XenApp. Although XenApp's Terminal Server legacy extends back to 1995, it never quite made it into the mainstream as a widespread desktop replacement. VDI, on the other hand, continues to generate extraordinary buzz and awareness. Whether View or XenDesktop, VDI is generally perceived as superior to the Terminal Services (now RDS) based model because of the elimination of concerns about session lock-downs, application incompatibilities and a different look and feel.

One of the big appeals of the virtual desktop model to IT administrators is its simplicity – it's just Windows running in a VM. It doesn't require specialists who know how to manipulate applications, dive into printing intricacies and work with registry hacks.

AltTech's Botti stated about the program, "I am excited about working closely with VMware, Lakeside Software, and Arrow financing and services, to provide a comprehensive way for users of Citrix XenApp to migrate to a full VMware View implementation. Providing the assessment, migration planning and financing to make this a nearly cost neutral process for the end user creates exciting new sales opportunities for our resellers."

 

Rough Chronology Behind AltTech's Virtual Desktop Trade-in Program

 

Alternative Technology founded	1986
Citrix founded	1989
Citrix WinFrame ships	1995
Alternative Technology signs distribution agreement with Citrix	1995
Citrix signs distribution agreement with Ingram-Micro	1996
VMware founded	1998
Bill Botti joins Alternative Technology as president & COO	2002
VMware coins VDI term	2005/2006
Alternative Technology signs distribution agreement with VMware	August 8, 2006
Citrix purchases XenSource	October 22, 2007
VMware ships 1st version of VDI: VDM 2.0.	February 1, 2008
Citrix ships XenDesktop VDI	May 20, 2008
Citrix discontinues distribution agreement with Alternative Technology	August 29, 2008
Citrix announces "Open Door" targeted to VMware customers	August 14, 2009
Alternative Technology Group of Arrow ECS announces Virtual Desk Trade-in program targeted to Citrix customers	January 11, 2010

Microsoft’s attempt to commoditize virtualization

"If I were VMware, I would be looking to lower my prices."

Laura DiDio, an analyst with ITIC. Reuters, July 6, 2009.

 

Both financial and industry analysts have berated VMware over the past 18 months for refusing to reduce prices despite Microsoft's commoditization of the hypervisor. VMware instead bumped the price of vSphere Enterprise Plus when it debuted last May, and yet continues to dominate the virtualization industry. The vast savings enabled from virtualizing a production data center make any delta in hypervisor costs a rounding error. Virtualization as a feature of the operating system is less than compelling to those organizations committed to an enterprise virtualization strategy. "Good enough" is not good enough.

I'm Not Cool Enough to be a VMware Person

A BrandPapers article defines commodities as, "…largely undifferentiated products that offer little or no perceived differences between competitive offerings." Maturing markets are more susceptible to commoditization which is signified by a uniform pricing level reflecting not much more than the marginal cost of production. Light bulbs, generic pharmaceuticals, mousetraps, computer RAM and coffee shop wireless service are some examples.

Microsoft knows a lot about commodities. It spends hundreds of millions of dollars defending its Windows brand against open source products. But it also takes the commoditization side as it did in 1996 by bundling Internet Explorer 3.0 as part of Windows 95 in order to crush Netscape. More recently, Microsoft launched a series of "laptop finder" commercials geared toward convincing would-be Mac purchasers that they pay an extra $500 for a logo. The Microsoft sponsored Endpoint Technology Associates white paper, What Price Cool?, claims that they pay an "Apple tax".

Particularly threatened by VMware's data center stronghold, Microsoft has been waging a vigorous campaign for the past year and a half alleging that not just basic hypervisors, but virtualization itself is essentially a commodity, stating "Virtualization is simply a role within the Windows operating environment." It misleadingly claims that an extra virtualization layer results in a "VMware tax". Its web site stresses the supposedly lower cost of Hyper-V virtualization including white papers, brochures, videos, charts, price-oriented case studies and a Microsoft vs. VMware Cost Comparison Calculator. Even the output of the Microsoft Virtualization ROI Calculator is 50% devoted to a cost comparison against VMware.

Other than price, Microsoft differentiates its virtualization solution by promoting the ability of Systems Manager Server Management Suite DataCenter (SMDS) to manage a heterogeneous environment incorporating multiple hypervisors and physical servers. This is only an attribute, however, in an organization approaching virtualization from a tactical, rather than strategic, perspective.

Hyper-V continues to rapidly improve as a product, and while not yet recommended by Burton Group for production environments, undoubtedly will be before long. Hyper-V R2 now includes the live migration feature ESX has had since 2003, but still lacks capabilities necessary for enterprise data center virtualization such as vNetwork. Cisco says this lapse results in the virtualization of 30% fewer servers while necessitating 30% more administrative time spent on the virtual network.

Evolutionary Approach to Virtualization

Microsoft has a vested interest in maintaining the large Windows-centric architecture common in most data centers and advocates a slow and limited approach to virtualization. Its white paper titled Microsoft Virtualization Delivers More Capabilities, Better Value than VMware states "…rather than undertaking a costly revolution, you should evolve your environment in a way that preserves and extends existing investments." In Microsoft's view, "physical machines…will continue to remain a key part of your infrastructure, as well as all the Hyper-V and VMware virtual machines."

The obvious disadvantage of a tactical, evolutionary approach to virtualization is the prolonging of high costs and inefficiencies along with unnecessary risk. A more insidious drawback is that layering virtualization onto an existing physical infrastructure inevitably leads to inefficiencies in the integration of compute, storage, network, backup and disaster recovery. The overall complexity of the data center increases along with the objects to manage which now include not only a physical architecture, but also virtualization hosts, virtual machines, hypervisors, virtual switches, virtual network adapters, virtual storage arrays, virtual back-ups, etc.

Achieving data center virtualization through an evolutionary process is also a difficult task. This was evidenced by Microsoft's Technical Case Study released early last year showing only a 50% internal data center virtualization rate with an expectation of "virtualizing at least 80% of new servers".

Why Adopt an Enterprise Virtualization Strategy?

The economics of virtualization tend to be extraordinary – it's not uncommon for an SMB organization virtualizing only 100 servers to realize millions of dollars in savings with an investment payback in fewer than 12 months. Larger organizations often have much higher savings and even shorter payback periods.

Virtualizing not just servers, but the entire data center provides a platform to unify technologies, equipment and processes. Desktop and disaster recovery silos become integrated components of a virtual infrastructure which in turn serves as the foundation for a comprehensive cloud computing strategy. Availability, recoverability, manageability, security and even performance can be improved beyond what is practical in the physical world.

VMware continues to thrive despite Microsoft's commoditization onslaught in part because of its enterprise virtualization features such as vNetworking, Fault Tolerance, Storage vMotion and Host Profiles. It has by far the most complete set of management and automation tools for a virtual infrastructure ranging from software development to disaster recovery to virtual machine lifecycle management. vSphere also includes published APIs that allow the leading manufacturers in security, storage and network to integrate their products directly into the hypervisor – resulting in still more efficient, scalable and secure environments.

Despite these many unique attributes, VMware's most compelling differentiator may be its astounding reliability. Unlike Hyper-V, it offers data center stability, performance and security that is independent from the bloat, reliability and patching issues of a general-purpose operating system. Even Redmond Magazine, "The Independent Voice of the Microsoft IT Community" gave its 2008 Editors Choice award for the most reliable IT technology to VMware ESX (the IBM mainframe came in #2).

The Importance of a Virtualization Focus

Microsoft's diversity in operating systems and other products gives it the financial leverage to offer virtualization as a free feature of Windows Server, but also detracts from an enterprise focus. Its home page, for example, mentions not a word about Hyper-V and instead promotes products such as Windows 7, Zune HD, Internet Explorer 8 and Office 2010.

VMware's corporate DNA is completely focused on virtualization as a strategic platform along with its role in enabling cloud computing. Its many years of experience and multiple levels of partner education and certification further differentiate its value from a commodity play. Denton County, Texas, for example, recently switched from a Hyper-V pilot to VMware vSphere for a complete virtualization deployment in part because of concerns about the lack of ISV support for Hyper-V. The county also found it difficult to find consultants with significant experience virtualizing data centers with Hyper-V.

Virtualization is a Lot More than Servers

In discussing VMware last July, Microsoft COO Kevin Turner stated, "We're a tremendous high-value, low-cost, high-volume software provider. That's our strategy, and that's what we're driving in this environment." This strategy has enabled Microsoft to become one of the most successful companies of all time and an incredible force in many different arenas. The enterprise virtualization market, though, is still young. Gartner says that only 16% of workloads run on virtual machines (eWeek 10/21/2009). A high-volume, low-cost product is far less important than the most reliable, secure and scalable solution.

A CIO advocating a data center transformation from physical to virtual may well put her job on the line. If successful, her organization will quickly realize huge savings along with other very significant benefits. It is imperative to choose the platform that best enables a complete, fast and successful transition. It is no time for compromise.

 

Author Disclosure: I work for a professional services company which is also a leading VMware partner, although we strive to provide the best virtualization solution for our clients.

Cisco UCS vs. HP Matrix: strategic vs. tactical approach to virtualization

"Revolutionary. Cutting edge. State of the art. These words and phrases are bandied around for so very many products in the IT field that they become useless, bland, expected. The truth is that truly revolutionary products are few and far between. That said, Cisco's Unified Computing System fits the bill."

Paul Venezia, ComputerWorld. November 10, 2009.

 

Following months of rumors about its "project California", Cisco made a big production last March in unveiling the Unified Computing System as transformative to the data center and "…as important to the industry as the personal computer was in the early '80s." Top executives from industry leaders such as Intel, EMC, VMware, Microsoft, Red Hat and others participated in the fanfare along with customers, partners and analysts. Cisco CEO, John Chambers, said, "This new Unified Computing System brings together the concepts of compute, network, virtualization and storage in a way that we think isn't just a product announcement, but we think is the future on which others will build".

Thirty-five days later, HP issued a press release announcing the HP Matrix with an obvious attempted jab at Cisco, "…the industry's first all-in-one software, server, storage and networking platform that allows customers to get the benefits of a converged system without a 'rip and replace' strategy for all their existing data center investments." HP's Web site and partner communications continue, albeit without much substance, to aggressively position Cisco UCS as inferior to Matrix.

An updated perspective is enabled by the several months that have now passed since both products began shipping. The overwhelming popularity, elegance, reliability and ease of deployment of UCS evidence the three years of investment Cisco made in developing an optimized virtualization platform. The complexity, limitations and lack of much excitement around Matrix, on the other hand, bear out a rushed repackaging of existing HP products in response to the Cisco announcement.

A Difference in Business Philosophy

Cisco approaches markets in terms of long-term architectural strategies rather than from the perspective of individual products or even product categories. Cisco foresaw the coming pervasiveness of data center virtualization and made partnership overtures to both HP and IBM years ago to to develop a more comprehensive computing architecture (Light Reading 12/9/2009). After being turned down by the server manufacturers, Cisco decided to undertake the effort on its own.

According to the Cisco book published early this year titled, Project California: A Data Center Virtualization Server, UCS is, "one of the largest endeavors ever attempted by Cisco". The Cisco-funded startup, Nuova, developed UCS under the leadership of VMware co-founder and former CTO, Ed Bugnion, to engineer a virtualization hosting platform for unifying the traditional data center functional silos of servers, storage and networking. Cisco UCS incorporates myriad innovations in architecture, performance, unified fabric and management.

The initial HP Matrix press release appears to be the first public mention of the product; it is hard to imagine that it resulted from a long-term data center strategy. The HP-sponsored April, 2009 IDC white paper, HP BladeSystem Matrix: Enabling Adaptive Infrastructure, says "HP is not introducing any brand-new technologies". Matrix not only lacks innovation, it feels like a work in progress. Even the "adaptive infrastructure" messaging used to introduce Matrix last April appears to have been replaced by "dynamic infrastructure".

HP Matrix – Time to Swallow the Red Pill

The HP BladeSystem Matrix Starter Kit includes the HP BladeSystem c7000 enclosure, HP VirtualConnect Flex-10 Ethernet modules, HP Virtual 8GB 24-Port Fibre Channel Modules, an HP Proliant BL460C (commonly referred to as the CMS – or Central Management Server) and 16 HP Insight Software packages. It also includes optional HP storage.

HP BladeSystem Matrix Starter Kit

• HP 10000 G2 Series rack
• HP BladeSystem c7000 enclosure
• HP Proliant BL460c commonly referred to as the CMS (Central Management Server)
• HP Virtual Connect Flex-10 Ethernet modules
• HP Virtual 8Gb 24-Port Fibre Channel modules
• HP StorageWorks 4400 Enterprise Virtual Array Starter Kit (optional)
• Infrastructure Operating Environment licenses (Insight software)
• Onsite BladeSystem Matrix Starter Kit Implementation service
• 14 existing HP Insight software and 2 new packages
  • HP Insight Software (IS) DVD - Integrated Installer 3.10
  • New HP Insight Recovery 1.00
  • New HP Insight Orchestration 1.0.2
  • HP Systems Insight Manager (HP SIM) 5.3.1
  • HP Insight Dynamics - VSE (ID-VSE) A4.1.2
  • HP Insight Power Manager (IPM) 2.0
  • Management Information Base (MIBs) 8.20
  • Performance Management Pack (PMP) 5.2.2
  • HP Insight Rapid Deployment software (RDP) 3.83.16 and 3.83.17
  • Remote Support Software Manager (RSSWM) 5.21
  • HP Insight Server Migration software for ProLiant (SMP)3.70
  • System Management Homepage (SMH) 3.00
  • HP Version Control Repository Manager (VCRM) 2.2.0
  • HP Virtual Connect Enterprise Manager (VCEM) 1.32
  • HP Virtual Machine Management (VMM) 3.7.1
  • WMI Mapper 2.60

HP BladeSystem c7000 enclosure

The version of the HP BladeSystem c7000 enclosure tested with, and certified for, Matrix runs only on single phase power which can cause some older data centers to require modification. Seven full height and a single half-height or 15 half-height device bays are available for populating with server blades although the first blade in the enclosure is reserved for the required CMS.

Central Management Server (CMS)

The Central Management Server is the brain of all operations and runs the Insight software. No fault tolerance or clustering and little to no redundancy results in a scenario of maintaining all eggs in one basket. The CMS requires either Windows Server 2003 R2 or SP2 / Windows Server 2008 R1. A failed CMS requires significant effort to restore it back on line. An unrecoverable CMS requires an on-site visit from an HP Engineering Services engineer along with a minimum one-week engagement to get the OS back on line.

Virtual Connect

A primary component of the HP BladeSystems Matrix, the Virtual Connect Flex10 module, provides some of the stateless capabilities that UCS enables along with a reduced cabling requirement. It has some negatives as well. For example, to move profiles between all the matrix-managed enclosures (chasses), all enclosures must be identical with the same number of VC modules in each enclosure, the same number of uplink cables all plugged to the same port numbers and with the same VLAN configuration an all enclosures. One enclosure cannot have more bandwidth requirements than other enclosures. Only four enclosures can be added to a single VC domain. Two chasses, each its own VC domain (the default), cannot be merged into one domain. One must be selected as the master and the other wiped out and then added to the existing domain.

Most importantly, the nature of Virtual Connect is to set up trunked ports between the core switch and VC modules for passing all VLAN traffic. This allows server teams to create additional LAN and SAN networks inside a VC domain and gives the server administrators control of the edge network. From a virtualized data center perspective, however, this scenario is disadvantageous in that it detracts from the networking team's responsibility for applying consistent network operations, policies and troubleshooting procedures. It is contrary to the joint efforts of Cisco and VMware in developing the VN-Link technology that enables the network team to effectively take back control of the vSwitch environment.

Insight Software

The HP Insight Software that comes packaged with Matrix is a collection of 14 pre-existing software packages and only 2 new packages. Page 8 of the HP BladeSystems Matrix Compatibility Chart shows the currently supported Managed node operating systems. Customers are locked into a small number that can be deployed, managed and controlled.

As the 33-page Insight Software Installation checklist guide shows, operating systems and databases are very limited. For example, Footnote 1 of Table 2.2 on page 11 warns that, "ID – VSE, VCEM, IO, and HP IR do not support CMS installation to 64-bit Windows Server 2008." Insight Orchestration requires Internet Explorer 6.0 SP3. CMS requires Microsoft SQL Server.

Overall Matrix options are quite limited as well. VMware vSphere, for instance, is supported in technology preview only. Matrix CMS only supports up to 250 logical servers whether they be physical or virtual servers (search for "250" on the HP BladeSystem Matrix Overview). While it is possible to combine multiple CMS units in order to reach an upper limit of 1,000 logical/physical servers, they are not clustered and do not share information. Server profiles cannot be moved from one CMS to another. This limits the Matrix as a solution to smaller organizations not wishing to utilize VDI, although VDI is impractical in any case since Matrix lacks automated provisioning support for desktop operating systems.

HP Matrix Implementation Service

Setting up all of the HP Matrix hardware and software is, not unexpectedly, very complex. A June 17, 2009 Infoworld review says, "The setup and initial configuration of the Matrix product is not for the faint of heart." The Matrix includes a mandatory two-week Onsite BladeSystem Matrix Starter Kit Implementation Service performed by a HP-Certified Matrix Professional from HP Engineering Services. But two weeks is still a short window for many organizations attempting to bring in all of the storage, network and server team players who need to provide input for the set-up – making implementations particularly challenging.

A customer cannot perform the Matrix install without a certified HP BladeSystems Matrix Engineer. HP also highly recommends HP Education Services for customer training and education along with Additional Technical Services. Matrix is QA/QC certified to only support a strict firmware, driver and server BIOS level. HP recommends that customers not update these components without first contacting the HP Matrix support line to ensure these updates will not negatively affect the overall Matrix infrastructure.

 

Cisco UCS vs. HP Matrix Matrix

	Cisco UCS	HP Matrix
Enterprise scalability	40 chasses, 320 blades – tens of thousands of VMs	250 total logical servers. Can combine up to 4 CMS to reach 1,000 logical servers, but no clustering or information sharing. Server profiles cannot be moved from one CMS to another
Redundancy	All components redundant	Central Management Server has no fault tolerance or clustering and little or no redundancy.
Memory	96GB Half Width Blade and 384GB Full Width Blade (8GB DIMMs)	With HP BL490C half-height blades : 144 GB w/8 GB DIMMs, 192 w/16 GB DIMMs 1 With HP BL685c (AMD) blades: 256 GB
"Closed" Architecture Limitations	Cisco UCS requires Cisco servers, CNAs and Fabric Interconnects for optimal performance	Requires one of the following specific HP ProLiant blades: HP ProLiant BL260c, HP ProLiant BL280c, HP ProLiant BL460c, HP ProLiant BL465c, HP ProLiant BL490c, HP ProLiant BL495c, HP ProLiant BL680c or HP ProLiant BL685c.2
vNIC & vHBA Support	Up to 128 each with Palo Adapter (56 vNICs per half-slot server today)	LAN – Ethernet 16 x 10 Gb downlinks to server ports SAN – Fiber 16 X 8 Gb 2/4/8Gb auto negotiating server ports
OS Support for Management Software	None required	Windows Server® 2008, Enterprise Edition 32 bit Windows Server® 2003, Enterprise Edition R2/SP2: 32 bit
Database Support for Management Software	None required	Microsoft SQL Server 2008, Microsoft SQL Server 2005 SP2, Microsoft SQL Server 2005 Express Edition SP2
Hypervisor Support	Supports any X86-based hypervisor. Particular advantages from tight integration with vSphere	VMware ESX Server 3.5.0 Update 4 VMware ESX Server 4.0 (pilot & test environments only) Windows Server 2008 Hyper-V (though not yet supported by Insight Recovery)
Guest OS Support (server)	Any	Windows Server® 2008, Datacenter Edition 32 bit and x64 Windows Server® 2008 Hyper-V, Datacenter1 x64 Windows Server® 2003, Enterprise Edition R2/SP2: 32 bit R2/SP2: x64 Red Hat Enterprise Linux 43 Update 7: 32 bit Update 7: AMD64 and Intel® EM64T Red Hat Enterprise Linux 53 Update 3: 32 bit Update 3: AMD64 and Intel® EM64T SUSE Linux Enterprise Server 103 SP2: 32 bit SP2: AMD64 and Intel® EM64T
Guest OS Support (VDI)	Any	None (No Matrix automated provisioning support )
3rd party development	XML-based API	None
QOS	Yes	None
Minimum cables required per chassis (inc. FC & redundancy)	2	6
Maximum cables potentially needed per chassis (inc. FC & redundancy)	8	34
FCoE	Yes	No
Ability to deliver native network and storage performance to VMs via hypervisor bypass	Yes	No
Network traffic monitoring & application of live-migration aware network and security policies	Cisco VN-Link / Nexus 1000V	None
Mfg. Support	1-Year	3-Year
Search.twitter.com hits: 12/15/09 – 12/22/09 in English [-ROIdude (my Matrix/UCS inquiries)]	54	12

¹The HP BL490c half-height blades support up to 144 GB with 8 GB DIMMs, or 192 GB with 16 GB DIMMs. They utilize 3 DIMMs per channel meaning that with over 96GB, the entire memory bus speed drops to 800 GHz. Additionally, the BL 490s have no RAID controllers and the SSD Hard drives are not hot-pluggable. Cisco UCS, on the other hand, uses a patented Cisco Extended Memory Technology which enables up to 384 GB on a full-width Intel Nehalem-based blade without sacrificing performance or requiring very expensive 16 GB DIMMs.

² While the c7000 will work with any HP ProLiant blade, Matrix only works with the blade models listed.

 

Cisco UCS – Revolutionizing Data Center Virtualization

Much has been written describing Cisco UCS both in this blog and in many other publications – obviating the need to go into details. At a high level, UCS is the culmination of years of development within Cisco/Nuova while also reflecting its close partnership with VMware. UCS was designed from the ground up as an optimized hosting platform for a virtualized data center. It integrates tightly with vSphere 4 to deliver an enterprise hosting platform enabling even the largest organizations to feel comfortable about virtualizing their data centers.

Despite its significant technology advances, Cisco UCS is surprisingly simple to install (HealthITGuy's Blog 11/25/2009 How Long Does it Take to Add UCS?). It utilizes only two (redundant) switches per 40 chasses and includes a common GUI making it easy for the server, network and storage teams to coordinate their efforts – yet be guided by role-based and resource-based management policies.

While Cisco UCS had already been running at IT infrastructure provider Savvis and other beta customers at the time of its announcement this past March, it appears that the first implementation of HP Matrix was at Stein Mart in the late summer. A December 9, 2009 BusinessWeek article reports that over 100 companies now are using UCS, and I suspect this is a quite conservative number. HP has not published Matrix sales figures, but supposedly a lack of qualified HP implementation engineering resources limit deployments to only a couple a month. If true, this means that Matrix production systems are likely somewhere in the 12 – 15 unit range.

Cisco UCS is a truly enterprise-class virtualization platform that, unlike traditional servers built for the physical world, is inspiring IT organizations with the confidence to embrace a completely virtualized data center strategy. This is not by accident. At Cisco's December 8, 2009 Financial Analyst Conference, Chambers said, "If you're reacting to what a competitor does, you're looking out the rearview mirror. You're three to five years behind." HP Matrix has a long way to go.

 

Author Disclosure: I work for a professional services company which is also a leading Cisco partner. I researched this article carefully, but welcome any corrective feedback.

Array-based backup advantages in a VMware virtual infrastructure

Eric Seibert wrote an excellent comparison earlier this week on searchdatabackup.com reviewing disk-to-disk backup applications for VMware: Veeam, Vizioncore, PHD Virtual and VMware VDR. As beneficial as these products can be, however, they have a disadvantage when competing against array manufacturer products such as NetApp's SMVI that are specifically designed for backing up virtual infrastructure.

The primary challenge faced by software backup solutions is their inability to offload impact from hosts, a capability that becomes imperative as VM consolidation ratios increase. Applications relying on host CPU cycles and disk IO to facilitate backup compete for those shared resources with the production workload. 

Most software backup tools rely on checksums of the entire VMDK to determine what data has changed and needs to be backed up – often a very IO intensive process that can even cause production applications to fail due to the resulting increased disk latency. According to Siebert's table, VMware VDR and Veaam take advantage of the vSphere vStorage APIs to utilize change block tracking. With the exception of Veeam running on a VMware Consolidated Backup (VCB) server utilizing change block tracking (CBT), software backup products require data deduplication, compressions and reads to transverse an ESX host − impacting the shared resources. And Veeam still requires a "copy-out" which produces far more IO than array-based snapshots.

Array-based data protection produces point-in-time copies without being IO intensive; it does not require walking the entire block set/file system/file structure every time a backup/replication occurs. Backup data can go directly to disk or tape without traversing the ESX hosts. Backups and restores consequently complete very quickly, and backups can take place more frequently throughout the day. NetApp's SMVI has yet another benefit in that it utilizes less storage since the footprint is deduped on both the production and backup datasets.

Utilizing different processes to protect and recover data for virtual machines, NAS content and any remaining physical servers – complicates not only backup, but also disaster recovery scenarios. Array-based solutions reduce complexity by eliminating the requirement to manage different replication schemes. Some of the array-based products further enhance disaster recovery by combining replication with VMware's Site Recovery Manager (SRM) to enable automated data center failover in the event of disaster.

The software backup manufacturers continue to add efficiencies and will undoubtedly increasingly take advantage of the vSphere vStorage API features to scale their products. Still, an off-board application/appliance will always be challenged to match the intelligence and efficiency of the storage array that owns the blocks (under VMFS) or file system (NFS). Array based data protection offers integrated backup and disaster recovery solutions that are both low maintenance and unified across the environment.

 

Russel Callen, INX Sr. Consulting Engineer, was a major contributor to this article.

Addressing concerns about the “newness” of Cisco UCS

Despite Cisco's recent entry into the "compute" data center sector, John Chambers said during the 10/24/2009 quarter earnings call that the UCS is experiencing, "…solid market reception with a very good initial ramp and order pipeline." As the first optimized hosting platform for a virtualized data center, Cisco Unified Computing System has generated tremendous industry buzz with many organizations eager to acquire it. Not unexpectedly, others are more cautious about moving to this new type of virtualization architecture. The following are some common IT staff concerns and how IT leaders can address them.

 

We know who to call when our Dell/IBM/HP servers break – our server team isn't used to the Cisco support model.

Virtualization changes everything. Server failure is no longer a crisis, or really even an inconvenience. Virtual machines simply move over to another host based upon preconfigured resource and proximity settings and, as software, quickly reboot – and users are back in business. Mission critical applications utilize vSphere's Fault Tolerance to migrate to another host without any session disruption at all. Cisco UCS includes exceptional redundancy within each chassis as well as between chasses. If a blade should break, it can simply be serviced when convenient. Additionally, Cisco provides outstanding support as validated by J.D. Power and Associates for the last three years in a row.

We don't need the fancy UCS capabilities of server profiles and stateless computing. VMware's High Availability, VMotion and DRS work just fine for us.

Service profiles and stateless computing may not be needed, but they sure are nice to have. If a blade should fail, a replacement blade can be configured in minutes through simply applying the desired service profile. Servers that must remain physical for regulatory or political reasons can still be rapidly recovered in the event of blade failure – the associated service profile is simply moved to a blade in a common sparing pool. Similarly, storage associated with a physical server can be replicated between Unified Computing Systems at different data centers, enabling very fast recovery in the event of a disaster. Server profiles also enable exceptionally fast provisioning as Cisco demonstrated by migrating over 150 of its servers to virtual machines in less than eight hours.

We are already taking on risk through virtualizing, and we are hesitant to add more risk with another new technology, Cisco UCS. What is the incentive?

A 2009 Webtorials study showed that server failure was the number one cause of outage, followed by human error. Virtualization, through high availability and vSphere Fault Tolerance, enables such fast recoverability that server failure is no longer an operational risk. The ability to provide all testing, patching and upgrades in a virtual sandbox lowers the risk of a production roll-out. Cisco Unified Computing System further reduces the human error component by utilizing a common GUI to assist coordination between the server, storage and network teams. The enterprise class performance and extended memory capabilities of UCS provide assurance for hosting even the most significant Tier 1 applications as virtual machines. Slashed cabling, switch requirements and lower on-going operating expenses provide further incentives for using Cisco UCS rather than traditional servers.

How do we know that Cisco is really committed to this new area?

Cisco has made a massive investment, perhaps a "bet-the-company" investment, in the UCS platform as part of its overall commitment to the data center. Cisco Unified Computing System is not simply a commodity server rolled off a factory belt; it is the culmination of an extraordinary effort and vision over several years to provide an optimized hosting platform for virtualized data centers. Cisco has risked disrupting its relationship with some of its biggest partners by entering this arena. Cisco's recently announced Acadia partnership with EMC and VMware further demonstrates its commitment to UCS and the virtualization space.

We are under-staffed, and we just don't have the spare time to learn how to configure and manage Cisco UCS.

UCS, like many new technologies, requires an initial investment in training in order to reap the huge rewards of reduced IT staff and resource requirements. UCS does not change the server, storage and network tasks, but helps simplify and coordinate them. Initial reports from the field show a very well thought-out, and not so intimidating, product.

We want to see other reference accounts in our area in our field. We don't want to be pioneers with arrows in our backs.

The UCS B-Series just started shipping within the past few months. While success stories are already starting to appear, it will take time to create and publish industry-specific references.

 

Cisco UCS C-Series

I recently tweeted a mnemonic tip to help keep the Cisco UCS series straight: B-Series stands for "Blades", and C-Series stands for "Can't believe Cisco Makes Servers". While tongue-in-cheek, I nevertheless was somewhat perplexed about why Cisco would enter the commodity market with a rack-mounted server, albeit one with significantly enhanced memory and performance capabilities. One reason may be that the product roadmap calls for C-Series to be managed by the same Cisco UCSM software that manages the B-Series. Organizations can purchase the C-Series today rather than standard servers yet preserve their investment once they make the full commitment to the B-Series virtualized hosting platform.

Denton County, TX migrates to VMware vSphere for a 100% virtualized data center

Industry analysts have long speculated that the release of Hyper-V R2 might cause organizations to consider shifting from VMware ESX to the Microsoft platform. The process of evaluating their hypervisor decision instead is leading some organizations to upgrade to vSphere and expand their virtualization environments. And in one recent case involving Texas' Denton County, it prompted a migration from Hyper-V to VMware vSphere.

"We knew that we wanted the whole virtualization enchilada," said Kevin Carr, Director of Information Services for Denton County. "But there are not a lot of consultants out there with experience virtualizing significant data centers with Hyper-V. The maturity and enterprise capabilities of vSphere made it a very interesting alternative."

Denton County initially only experimented with virtualization, and as a Microsoft shop familiar with Windows, decided to utilize Hyper-V as its virtualization platform. While starting with only a few test VMs, new servers began to be put into production as virtual machines until the County reached a total of 24 VMs running on four Hyper-V hosts. As the number of VMs continued to grow, IT management increasingly became aware of virtualization's capabilities to lower costs, reduce power usage and increase availability. They grew interested in the idea of virtualizing their remaining 139 physical servers, and were looking forward to the increased capabilities that Hyper-V R2 would offer.

Carr said that his team began to have some initial concerns about using Hyper-V as their platform when they discovered that several of their ISVs had statements supporting VMware, but not mentioning the Microsoft hypervisor. After speaking with their INX account manager, they began evaluating VMware vSphere as an alternative and found it to be very tempting. They saw vSphere's Fault Tolerance for mission-critical servers as a superior option to clustering software and hardware solutions. They also liked vSphere's power management capabilities which would enable them, as a primarily 8-5 operation, to automatically consolidate and shut down unused servers – thereby saving both power and cooling costs.

"As a big Cisco shop, we wanted the ability to manage both the virtual and physical networks with the same Cisco interface with which we are familiar," said Carr. "The Cisco Nexus 1000V virtual switch, in conjunction with vSphere, provides that capability while also reducing the amount of time required to manage the virtual switches by providing an aggregated management console."

The other vSphere management capabilities were also appealing according to Carr. His team liked the single pane of glass enabled by vSphere vs. the multiple consoles required to effectively manage a Hyper-V environment. The county currently is in the process of deploying nine new Dell R710 rack-mounted servers as the hosting platform for supporting a completely virtualized data center running on VMware vSphere Enterprise Plus.

The antidote to virtualization politics? IT leadership

  
Virtualization as an enterprise data center strategy is transformative, yet continues primarily to be deployed as a point solution – driven by projects such as ERP implementations, hardware refreshes or disaster recovery initiatives. This approach risks virtualization becoming yet another technology island – increasing overall complexity even as it reduces select capital and operating expenses. Adversity, though, provides IT leaders with an opportunity. They can slash IT costs, empower their staffs and promote their own careers by taking up the mantle for a fully virtualized data center standard.

The Need for IT Leadership

Despite virtualization's huge buzz, both IDC and Forrester estimate only a 30% enterprise virtualization penetration rate, while Gartner says it's even lower. Lack of application manufacturer support, regulatory prohibitions and security concerns are common roadblocks, but the biggest inhibitor is inertia. A fully virtualized data center requires both vision and the ability to execute upon that vision – overcoming the inevitable political hurdles in the process.

In the absence of leadership, virtualization tends to work its way slowly through the data center in a piecemeal fashion, requiring increased administration even as the cost for facilities and equipment decline. Administrative requirements are compounded by the necessity of managing both virtual and physical environments.

Lack of an enterprise vision perpetuates myopic decision-making. The inevitable hodgepodge of technology islands is antithetic to a virtualized data center dependent upon a cohesive architecture spanning compute, storage and network. The risk of dissipating huge virtualization savings makes "good enough" compromises not good enough. Basing a hypervisor decision upon licensing costs, for example, disregards VMware vSphere's singular ability to enable high percentage virtualization of the data center, particularly when augmented with the Cisco Nexus 1000V virtual switch.

Equipment decisions similarly should be driven by their contribution to data center virtualization success rather than by traditional brand loyalties. Pooled compute resources, for instance, make individual servers interchangeable and non mission-critical commodities. On the other hand, Cisco's revolutionary Unified Computing System (UCS) provides such a superior platform for hosting virtual infrastructure that IT leaders do their organizations a disservice by not at least investigating its myriad advantages.

The capabilities and efficiencies enabled by enterprise virtualization provide an effective means of challenging the status quo. An application owner may, for example, insist that SQL Server be run in a clustered physical environment because that's the way it's always been done. The IT leader can counter by calculating the huge monies this predilection will cost the organization versus using VMware vSphere's Fault Tolerance.

Creativity overcomes most obstacles to a fully virtualized data center such as the minority of software manufacturers still refusing to support their applications on VMs. An easy resolution for organizations running vSphere over NFS on NetApp is to FlexClone a Virtual Machine's virtual disk into a LUN and simply present to a physical server. Voila, instantaneous V2P.

Sometimes, as when constrained by regulatory mandates, servers must remain physical. Innovative technologies such as PlateSpin PowerConvert enable both compliance and virtualization advantages of enhanced testing, pseudo high-availability and DR replication by continuously snapping a physical server to a VM. The ROI takes a small hit, but a 95% virtualized data center is still vastly superior to a 30% virtualized one.

Roadmap to a Successful Virtualized Data Center

The journey starts with building a compelling business case. Expanding from partial to full data center virtualization enables a very large and measurable ROI, but the business value hardly stops there. A 2009 Webtorials Business Continuity report says that server failure is the number one cause of outages – a problem eliminated by enterprise virtualization. Green initiatives, cloud computing benefits, vastly enhanced disaster recovery, reduced IT staff time and more productive virtual desktops can be incorporated, along with a comprehensive ROI analysis, into a virtualization justification document that is presented to senior management and the Board.

As with any enterprise project, virtualization planning is essential. A virtualized data center not only impacts all IT infrastructure groups, but also demands much higher levels of coordination between them. This warrants representation from each IT function in the planning phase as part of a "data center team". Under the IT leader guidance, they can ensure that the backup, storage, network and security components are designed and integrated as part of a cohesive architecture. They can also anticipate required IT organizational changes – incorporating tools and processes for managing new virtualization challenges such as limiting server growth no longer constrained by budget and procurement delays.

Employees, management and customers enjoy success stories. IT leaders can build organizational pride by publicizing their virtualization project – certainly internally and perhaps externally as well. CO2 emission reduction can be touted – potentially along with rebates offered by many utilities to customers that virtualize, as part of environmental initiatives. A promotion is not an unusual accompaniment to a successful enterprise virtualization deployment.

The desktops may be virtual, but the ROI is real

Fielding questions while on the VDI panel at last month's Vail Pacific Crest Forum, it became apparent that investment analysts consider a lack of VDI ROI to be common knowledge. But while the savings resulting from desktop virtualization may be less obvious than with server consolidation, they nonetheless are often substantial.

Gartner's March 2009 press release predicts exploding VDI sales of $65B by 2013 – equating to 40% of the professional PC market, and up from only $1.4B and less than 1% today. This represents "…part of bigger shift in client computing from traditional thick-client distributed PCs toward more manageable, secure and centralized client computing environments." A VMware-sponsored IDC analysis of View customers indicates that a centralized VDI approach produces measurable efficiencies.

The IDC white paper shows that compared with using unmanaged PCs, organizations deploying VMware View saved an average of $287 per user annually from lower device costs and $601 from lower IT support costs. Another $130 was saved from improved user productivity resulting from faster start-up time, less security issues, etc. The overall 5 year ROI was 367% with a 5.61 month payback.

While the white paper lacks supporting data, the numbers nonetheless look reasonable. For comparison, I recently calculated annual savings of $455 for an organization virtualizing 1,000 PCs and laptops as part of a phase one View 3 deployment. The payback period of 11.7 months against an investment of $500,000 is in the general vicinity of the IDC averages. Applying the IDC white paper estimate of $130 in user productivity savings further reduces the payback period to 9.3 months.

The only support savings calculated are those specifically related to hardware issues. The organization's extraordinarily well-managed physical desktop environment includes extensive tracking and categorization of support request – conferring unusual confidence in the numbers.  

 

The analysis assumes that as PCs and laptops reach their refresh periods of 48 months and 36 months respectively, they are replaced with Pano Logic zero clients. The already mostly virtualized data center reduces the incremental storage, server and services costs from those required for a green field VDI deployment.

Annual HW/SW savings of $131 per user result primarily from no longer needing to refresh the $800 PCs and $1,200 laptops (including tax, shipping, set-up, etc.). The $324 in annual hardware-related support costs per PC/laptop does not include typical costs of less well-managed environments such as OS and application patching/upgrades, desktop imaging, etc.

This organization lacks small remote offices that could significantly contribute to VDI savings by eliminating local network infrastructures. And the ROI doesn't quantify tangential VDI enhancements such as much faster boot times, greater stability, reduced security risk and ubiquitous desktop access from anywhere users can get to a browser. It also disregards significant disaster recovery advantages.

Embracing VDI

Tyler Rohrer makes a case on www.vdi.com that VDI is not about reduced Capex or Opex, but about increased potential user productivity. It certainly is useful to compute not only productivity benefits, but also to evaluate the positive impact of virtual desktops on organizational objectives such as "green IT", employee empowerment and business process efficiency. But at the end of the day, CFOs expect to see hard metrics for VDI savings. Fortunately, it frequently is possible to show a compelling ROI.

Note: This article is also posted on http://www.dabcc.com/article.aspx?id=12094

 

VMworld UCS sign brings back an unrelated memory

The sign at the entrance to Moscone Center last week specified that the impressive display of 16 UCS systems was not just "blinking lights", but was powering VMworld. That made me think of a situation I was fortunate to escape where racks of mostly blinking lights were used to fraudulently extract $120 million in loans from unsuspecting banks.

On September 12, CNBC's American Greed will air Fraud in Cyberspace, which chronicles the career and Ponzi-like scam of Barton Watson, CEO of the CyberNET Group. He used proceeds from newer borrowings to stay current with older loans while continuing to fund his lavish lifestyle. When the Feds finally stopped the music, he was found to owe somewhere north of $100 million against assets of only $2 million.

I first encountered Barton, or "Mr. Watson" as his employees were forced to refer to him, nine years ago when he came to northern California to discuss acquiring my Citrix integration company, RYNO Technology. Barton had been recommended to me by a senior Wyse VP, and seemed to have impeccable credentials including a Stanford MBA and former Goldman Sachs partnership. During the next six months, we had over a dozen meetings including multiple all-day affairs at his opulent Grand Rapids, Michigan headquarters.

Certainly, the first thing most people would notice about Barton was his sheer size of around 6'2" and 400 pounds. This would be quickly followed, though, by an impression of extraordinary intelligence. Barton had an uncanny knowledge about almost any subject. A valedictorian of his Belding, Michigan high school, Barton was said to be without question the smartest kid in the school district.

Barton and his wife, Krista, traveled first class around the world staying in the best hotels and eating at the finest restaurants. He drove a Ferrari, Bentley and Rolls Royce among other luxury autos. While I've never been much for the trappings of luxury, I was beguiled by Barton's charm and intelligence. He and Krista would frequently fly to San Francisco where he'd meet my wife and me for wonderfully entertaining meals at expensive restaurants. I met his top executives and spoke with his two largest Citrix clients, both of which gave his organization glowing references.

I began to see another side of Barton as negotiations progressed. He exploited his talent of quickly reading character in order to throw people off balance and make them more susceptible to his thinking. The "gut feeling" cliché became increasingly tangible for me in the form of an aching stomach, but I was determined to see the acquisition through… until he made an offhanded comment about being friends with Citrix CEO, Mark Templeton. Shortly afterward, I happened to run into Mark who assured me that he did not know Barton…and Barton was not the type of man anyone was likely to forget meeting even one time.

This encounter prompted me to hire a private eye firm to investigate Barton which eventually revealed a significant record of litigation, fraud and jail time. In the interim, I did some deeper due diligence on my own. I called back CyberNET's two big Citrix references, and this time reached someone different at one since Barton's contact was on vacation. The replacement said terrible things about CyberNET and that his company would never again work with them. I also called up both Stanford University and Goldman Sachs to verify Barton's claims of an MBA and former partnership respectively…both of which turned out to be untrue.

I emailed Barton about the strange inconsistencies and he angrily retorted, "I will not have my integrity questioned. THE DEAL IS OFF!!!" This was quickly followed by a letter from his attorney demanding the return of every document related to our aborted merger and stating that if I mentioned a word about the details of our transaction to anyone that I would be sued.

Barton managed to keep his charade going another four years. CyberNET was listed as #110 on the VARBusiness 500 with 2003 sales of $300M, though actual sales estimates are between $10M - $12M. On November 18, 2003 the Feds surrounded Barton in his posh Grand Rapids, Michigan home. After drinking a $700 bottle of Le Tache wine, he placed a 911 call to police. "Hi, this is 911 Emergency, what's your emergency?" asked the sheriff's dispatcher. "Well, there's a gun in my mouth," replied Barton. He repeatedly urged the dispatcher to relax. "The theory is that I wouldn't call you if I didn't want help. But…I just want someone to clean up the mess." Barton proceeded to lecture the dispatcher about what was necessary to make a good bottle of wine. He was found dead at the age of 44 from a gunshot wound to the head.

Excerpts from my VMworld 2009 vExpert presentation

I was surprised when John Troyer asked me to present on evangelism to the group of charter vExperts. I mean, come on...that's like being asked to speak about partying to a San Diego State frat house. But being reluctant to ever turn down an opportunity for exposure, I of course accepted.

---

While being fortunate that I don't have to justify my daily activities, I do want to ensure that senior management knows they're getting a good return for their money, and that my blogging and Tweeting is actually benefiting the firm. This is more important now that other company execs are starting to follow me on Twitter and can see me Tweeting during the middle of the day instead of doing real work.

---

Differentiation works personally as well as professionally. Seven years ago I found myself in a surprising divorce. I dreaded the prospect of going back into the dating scene...I was never that successful the first time around, and now I was a heck of a lot older. But I quickly discovered Internet dating, and then was pleasantly surprised to learn that most men have little concept of marketing. Most match.com tag lines read something like, "Great catch", "Last good man on earth", or "Likes to take long walks on the beach at sunset". I, naturally, took a different approach.

My dating experience was vastly better the second time around, and I eventually met a great lady who I married. I've drafted her as part of my evangelistic efforts – she assisted Krystal Lowe yesterday by driving the VMworld spouses up to tour our good customer, Korbel Wineries. By the way, the VMworld wives were joking about how their geeky husbands upon arriving at the Moscone Center immediately flocked to the huge Cisco UCS display to take pictures.

---

John asked me whether or not being really smart is necessary to be a good evangelist. I can assure you from first-hand experience, it is not. Over the years I've co-authored many Citrix books and a VMware book. Many people assume that I'm an engineer who knows what he's talking about. The reality is that I work with very bright co-authors who are far more knowledgeable than me, but I get to share in the glory.

---

Speaking with a client led me to the idea of approaching PG&E about extending rebates to clients who virtualize. I told a couple of our VMware reps about it in confidence, but otherwise kept the program secret thinking that it was a competitive advantage for AccessFlow. But then Diane Green was on the stage at VMworld along with the CEO of PG&E, never mentioning AccessFlow or Steve Kaplan. It taught me a good lesson about the folly of trying to horde good ideas. I might as well put them out there and get credit for them.

---

The first company in the world to receive a utility rebate check for virtualizing was our good customer, 1-800RADIATOR – which has about 25% of the U.S. radiator market. The CTO was speaking for us at a VMware seminar, and held up our T-shirt which had a picture of VirtualMan unplugging a server and which read, Fight global warming…virtualize now. The CTO said, "Steve gave me this T-shirt and I'm thinking, 'What the heck is this? We're a radiator company. We like global warming.'"

---

Like many things, you can either put a little effort or a lot of effort into evangelism. If it's the latter, I encourage you to do some self-reflection in order to ensure that your personal objectives are in line with the level of time commitment required. In my own case, before I decided to commit to a high level of evangelism at INX, I first did some extensive soul searching. What I realized was that, deep down, what I really want….is to make a lot more money!

 

 

Random vmworld 2009 thoughts

There's no recession at San Francisco's Moscone Center. With a reputed 13,000+ attendees, the exhibit hall was packed and the sessions sold out. The energy is almost palpable and a nice break from the past year of continual gloomy economic news. Citrix added to the fun by postering virtualization advertisements on taxis and benches around the Moscone Center.

Living in the east bay, I only decided to stay in San Francisco for VMworld around a month ago. I ended up at the Union Square Hotel. It's nice enough, but I'm on floor 2 ½. Really. I feel like I'm in "Being John Malkovich". The work-out facilities turned out to be at a sister hotel a few blocks away, so I figured I might as well just get up early for Bikram yoga, and was at class at 6:00 a.m. so that I could still catch early morning meetings.

My favorite part of the Partner Day general session was VMware's new messaging around VDI: User-centric. As a Citrix channel partner for many years, we had to regularly adopt new messaging: thin-client, SBC, access infrastructure, on-demand computing, application delivery. I think that "user-centric" computing is a great summation of what VDI needs to accomplish in order to become the new desktop standard.

On the part of INX, our team did an amazing job at bringing our Extend2Cloud™ demo together for VMworld. Extend2Cloud™ utilizes VMware vSphere, Cisco Nexus 1000V, Cisco UCS, AFORE and our own virtual appliance to extend fibre channel through the WAN and present storage in geographically disparate data centers as if it were one logical data center irrespective of distance. Tonight we demonstrated vMotion between two data centers 60 km apart. Tomorrow we will hopefully be able to demonstrate vMotioning to Europe.

I met with the CEO and team from Hyper9 this morning and was impressed with the product's management capabilities. I'm eager for our engineering team to check it out. I also finally met Tyler Rohrer "t-rex' of Liquidware Labs in person. Our representative at the VMware Partner Technical Advisory Board said that the product was highly touted at the PTAB meeting today. Liquidware was also clever enough to purchase the vdi.com domain and launch a well-done VDI-focused social Web site.

Among the highlights for tomorrow - lunch with Doug Brown of DABCC. I always enjoy the opportunity to chat with Doug. He may espouse an "average Iowa boy" persona, but as his hordes of readers know, Doug is an exceptionally astute observer of our industry.   

 

 

 

 

Intel’s Nehalem virtualization game-changer

A client recently purchased a 4-CPU Dell R900 to expand its VMware environment rather than the 2-CPU Dell R710 because the IT staff felt it would scale better. I hope it's not too late for them to return the server.

The R710 may have only two CPUs, but these are the 11 Generation Intel Xeon 5570 CPUs. The revolutionary Intel 5500 series CPU (Code named Nehalem) was designed to optimize virtualization performance and includes Intel Virtualization Technology for Directed I/O, QuickPath Interconnect, Integrated Memory Controllers, Hyper-Threading, hardware-based memory protection, Extended Page Table, and much more.

Dell's Web site says that the "2-socket PowerEdge R710 achieved the best performance for virtualization, better than all other 2-AND 4-socket servers." This claim is based upon VMmark scores which show a 2-CPU (8-Core) R710 handling slightly more VMs than a 4-CPU (24-Core Xenon 7460) R900. (While the R900 was running ESX 3.5 rather than vSphere for the R710, the results were similar to the AMD based 4-CPU, 16-Core R905 running vSphere).

Dell's server configuration tool shows a cost of $17,203 for a vSphere-ready R710 with 96GB of RAM versus a purchase price of $26,173 for a similarly configured R900 with four 6-Core CPUs. While the R900 scales to 256GB vs. only 144GB for the R710, this adds $19,113 of cost versus $966 to scale the R710 to 128GB. If 256GB is required, it makes more sense to simply purchase a second R710 (or Cisco UCS – see "Alternatives to Dell" below). Another advantage – the R710 uses only 2U of rack space vs. 4U for the R900.

The purchase cost disparity, advanced capabilities and reduced rack space requirements make the R710 the obvious choice as a virtualization host. But using only two CPUs provides further software licensing benefits. Here is a recap comparing three year licensing costs assuming the host is running both VMware vSphere Enterprise Plus and Cisco Nexus 1000V along with both Microsoft Windows Server Data Center Edition and SQL Server Enterprise Edition. The Microsoft products are both licensed according to the physical host CPUs and allow an unlimited number of the virtualized instances of the software to run on the host.

 

Alternatives to Dell

HP and IBM also offer Intel 5500 based servers. If the virtualization environment is sizable, however, we recommend evaluating the Cisco UCS as an optimized virtual infrastructure hosting platform. The superior UCS performance and unified fabric capabilities can further reduce licensing expenses along with rack space, power and cabling costs. Its integrated management capabilities help drive complexity out of operating the environment.

Riddle me this: When is a server no longer a server?

 

When my wife refers to her "dad", her siblings know whether she's talking about her biological father or stepfather. Virtualization consultants similarly tend to have little trouble differentiating whether a "server" refers to a virtual machine or to something upon which a virtual machine is residing. But Cisco's Unified Computing System (UCS), as the first specially built virtual infrastructure hosting platform, is confusing even industry experts.

Just one recent article, (August 1, 2009 Redmond Mag.com) reports that Chris Wolf of Burton Group, "…believes Cisco will fight an uphill battle in the competitive blade server market" while in contrast IDC says that UCS, "offers a clean-sheet approach to solve data center challenges". Greg Schulz of The StorageIO Group, observes that UCS is, "…an integrated platform that offers all data center building blocks in a unified form factor.", yet also remarks that, "…Cisco is new to the server world." Staten and Schreck of Forrester Research state that UCS "…is the next step in blade server technology" but also claim that, "No one is clamoring for another server vendor."

Historically it hasn't mattered if organizations were virtualizing 100 or 1,000 servers; they have only had one choice for hosting those servers – a server. The problem is that servers were designed to run applications, not virtual machines. This has led to myriad challenges ranging from lack of enough memory and network ports to cabling issues to resource provisioning frustrations.

UCS is creating a huge industry buzz because it was developed over a period of three years to incorporate the performance and management requirements necessary to optimize virtual infrastructure – and IT staffs are quickly grasping its game-changing capabilities. The synergies between UCS and VMware vSphere extend well beyond the technical integration. Just as an enterprise vSphere deployment creates a compelling opportunity for UCS, the Cisco platform in turn generates a higher level of comfort for virtualizing 100% of the data center.

Server Semantics

An easy solution to the terminology turmoil is to simply refer to UCS, or to anything hosting VMs as, well, a host. This makes it easy to both categorize and understand UCS. A traditional server can, of course, also be a host – but when it's a host it's no longer a server. 

Now that the server semantics are settled, we can turn our attention to other philosophical ruminations such as, what exactly is a desktop?

VMware's got plenty of mojo

Randall Kennedy claims in his 08/11/2009 InfoWorld article that VMware is, "...a firm that has lost its confidence, its stride, and its mojo." He says that VMware looks, "like a middle-aged has-been going through the motions."

Has-been?

VMware's $420 million acquisition of SpringSource is hardly the type of bold move indicative of an organization that is, as Kennedy contends, "a market leader in decline." And while Microsoft, Citrix and Oracle are much older manufactures who fiercely compete against VMware with less expensive hypervisors, Gartner says VMware still dominates the industry with a 90% market share.

VMware's mojo appears very well intact judging by the 14,000 customers and partners who will inundate San Francisco later this month for only the sixth U.S. based VMworld. Compare this with the attendance of the eleventh U.S. based customer event of another great software company, Citrix Systems. Synergy 2009 had a record 3,069 attendees.

Kennedy's repeated insistence that that VMware has become complacent is remarkable in the face of its recent launch of vSphere, the largest announcement in the company's history. The development of vSphere encompassed over 3 million engineering hours during a 3 year period by over 1,000 engineers.

Although Kennedy contends that VMware lacks "real innovation", vSphere incorporates remarkable advances in compute, storage, network, security and management. But vSphere is much more than the sum of its 150+ new features – it fulfills the performance, reliability, management and security requirements to establish virtualization as the standard and the foundation of a 100% virtualized data center.

As expected of a dynamic industry leader, VMware continues to garner exceptional industry support. New products developed by well-known manufactures including Symantec, McAfee, Check Point, EMC, NetApp and many others integrate with vSphere's published security or storage APIs. The vSphere launch in Palo Alto was attended by a long list of industry luminaries including John Chambers who personally took the stage to affirm the commonalities of the Cisco and VMware virtualization visions.

Virtalization esoterica

Other than brief complaints about ACE and ThinApp, Kennedy's argument is based upon his belief that a leaked list of VMware Workstation 7 features shows a lack of "game-changing innovations". Even if true, it is ridiculous to extrapolate, as Kennedy does, supposed shortfalls in this sliver of VMware business to its dominating core products and corporate culture.

Kennedy maintains that VMware's response to competition is, "…retreating further into the ever-shrinking technical superiority niche currently occupied by VMotion and other virtualization esoterica". Esoterica? While VMware and Citrix already include it, Microsoft has long recognized the importance of live migration. By far the most anticipated feature of the October launch of Hyper-V R2, Microsoft's live migration innovation would be more impressive if VMware hadn't first introduced VMotion six years ago.

 

To “V” or not to “V” – the economics of 100% data center virtualization

Data center virtualization still has a long way to go. Gartner estimates global virtualization penetration to hit 20% this year. But now that the combination of VMware vSphere and the Intel Nehalem chip (5500 series) enable 100% virtualization, controversy is swirling around whether or not organizations are ready to go there. One well-known technology magazine editor recently sent me the following email:

"I disagree with the statement that IT managers are looking for 100% virtualized infrastructures. It's actually quite the opposite. Many large companies with thousands of applications have divided their server farms into targets for virtualization with strict guidelines on what will remain on physical servers. Maybe smaller shops will eventually be 100% virtual, but you're looking at a 10 year roadmap for the majority of companies to get there."

Many IT industry players share this opinion. A recently circulated IDC white paper from last year, for instance, defines advanced virtualization as "25% server penetration and at least some storage virtualization". Organizations are nonetheless increasingly virtualizing their entire data centers. Examining the economics quickly reveals why.

SampleCo Virtualization Platform

SampleCo has virtualized 50% of its servers – the virtualization penetration percentage that IDC projects within the next year. The low-hanging fruit of 110 test/development servers and lightly used (non-mission critical) production servers are VMs, while 110 production physical servers remain. A 5-year savings projection from completing SampleCo's data center virtualization follows along with the required investment.

With an ROI of 512% and only a 9.8 month payback, it clearly makes sense to complete the data center virtualization even if these numbers are widely off the mark. And the ROI only includes certain easily quantified areas of cost savings. Other common benefits include:

• Staff savings
• Potential elimination of planned data center expenditures such as PDUs, A/C units, expansion, etc.
• SQL Server license savings (this can easily amount to hundreds of thousands of dollars)
• Windows Server license savings (this number can be significant depending upon the licensing needs)
• Elimination of server downtime
• Significantly enhanced DR capabilities
• Laying the virtualization foundation for embracing cloud computing
• Potential for improved security and management
• Establishing the infrastructure to facilitate a virtualized desktop environment – and a lot more savings

5-Year SampleCo Savings

Servers

Even higher-end production servers generally cost under the $10,000 figure used, but it includes costs such as shipping, tax, administrative time, rack & stacking, rack space, cabling costs, SAN switch ports, network core switch ports, network distributed costs, share of the UPS and generator costs, etc. We assume a 5-year server refresh cycle meaning that all of the existing servers will be upgraded sometime in the 5-year ROI analysis period.

vSphere Hosts

We will refresh the vSphere hosts according to manufacturer best practices of 3 years, and therefore include replacement costs for the 10 vSphere hosts.

New Servers

The physical scenario assumes SampleCo continues purchasing roughly 10% of its physical server base annually – or around 10 servers per year. Purchase of additional ESX hosts is not required because as the existing units are refreshed, the replacement models will handle far more virtual machines.

Power/Cooling Servers

We did extensive analysis to come up with a $75 monthly utility cost for powering a server. This figure also reflects the cost of cooling a server, which can be 2 – 3 times higher than the cost to power it depending upon the efficiency of the data center. It also reflects allocated portions of power for the switches, UPS, generator, etc.

Power/Cooling New Servers

If SampleCo continued purchasing physical servers, we assume that on average – half will be on-line for the ROI analysis period.

Maintenance

Since the 110 remaining servers are the more mission critical, we assume they are on a $1,000 annual maintenance contract for each of the two years past the manufacturer 3-year warranty expiration.

SampleCo Virtualization Expansion Investment

Despite the buzz around the hypervisor becoming a commodity, today we are at the stage where it is imperative that we choose the best virtualization platform in order to enable 100% data center virtualization. Although Burton Group recently included Citrix XenServer 5.5 plus Citrix Essentials 5.5 Platinum Edition as one of the two enterprise production-ready virtualization platforms, Only VMware vSphere Enterprise Plus is used in the ROI model for the following reasons:

1. VMware vSphere Enterprise Plus is the only hypervisor today that resolves the network concerns around the inability to effectively establish network and security policy at the VM level, enable policy live migration, isolate a VM at the physical port level for troubleshooting, and control and audit VMs as they move between physical hosts. Additionally, vSphere support of the Cisco Nexus 1000V enables the network administrators to maintain configuration responsibility for the vSwitches with aggregated management of the vSphere hosts. And they can do so using familiar Cisco tool sets.
2. VMware vSphere has more enterprise capabilities than any other hypervisor such as storage vMotion, management, security and storage APIs, hardware hot add, Chargeback, etc. It integrates both with additional enterprise products such as Site Recovery Manager and Lab Manager as well as with a huge array of third-party products.
3. VMware vSphere enables more VMs per host, meaning that the associated cost of hardware, power, rack space, network ports, etc. is reduced – making vSphere the most cost-effective solution for a 100% virtualized environment.
4. While Microsoft is shooting for a 50% internally virtualized IT environment, VMware is already at 100%.
5. Gartner says that VMware's market share is 90% - making it reasonable to assume that our sample organization is already running VMware and will want to continue with the platform.

   Storage

Many factors go into estimating the storage required to virtualize the 110 new servers. For instance, as larger servers – many of them will likely already be SAN attached. The existing SAN, the type of storage and available capacity all affect projected storage costs. Using the thin-provisioning features of vSphere for the existing hosts will likely free up capacity as well. For purposes of the ROI, we use a cost of $50,000 which is both the figure our experienced engineers estimated, and the rough number calculated by assuming 100GB of added storage at $5/GB for each of 110 of the new VMs.

Hosts

We assume rack-mounted servers rather than the Cisco UCS (which is an option we certainly evaluate in a real deployment). Ron Oglesby, Virtualization Practice Executive for Dell, gave a presentation at 2009 Virtualization Congress showing that the Dell R710 (with two Intel 5500 Series CPUs) handles 24 VMs – more than twice the virtual machines as the 7400 Series R910. Since the 110 remaining servers to virtualize are the more resource-intensive machines, we'll use a more conservative ratio of 12 VMs per R710 host. This means that we'll need ten 2-CPU servers including redundancy. At a street price of around $13,500 per R710 with 96GB, we'll round up to $15,000.

Vsphere & Nexus 1000V licensing

While the Nexus 1000V is not required and doesn't directly impact ROI savings, the cost is included because of its pivotal role in enabling 100% virtualization. At a street price of around $3,600 per CPU for the vSphere 4 Enterprise Plus/Cisco Nexus 1000V bundle, SampleCo will need to purchase $72,000 of software along with an annual Platinum Subscription & Support averaging around $18,000 per year.

IT Staff Training

Since the staff is already trained on ESX and very comfortable with the technology, we'll just send a few staff members to a 2-day vSphere 4 training class ($1,050).

Professional Services Implementation

This figure will vary widely depending upon the consultants utilized and the amount of work the internal staff does, We assume a cost of $50,000 for assistance with assessing the existing physical environment and planning the vSphere/Nexus implementation including networking and security best practices.

Will UCS Unify IT Staffs?

This article also is available on www.dabcc.com

Traditional data centers tend to be segregated into functional silos - a model that has endured in the physical world largely because of the crisp lines of demarcation. Virtualization is shaking up the status quo and increasinng demand for more efficient IT processes. Cisco's UCS helps automate and streamline the stovepipe model by unifying network, storage and servers with virtualization. It also provides an opportunity for IT organizations to begin chipping away at those silo walls.

 

Dysfunctional Computing

New IT infrastructure tends to be project driven. Large application implementations and other departmental endeavors provide budget for servers and SANs – filling up data centers with a hodgepodge of technology islands in the process. Gartner's number one energy saving recommendation at its 2007 Data Center Conference was to turn off servers that appear idle and see if anyone complains (searchcio.com: Top 10 ways to save energy in the data center).

Function tends to follow form with technology specialists typically managing their own domains in relative isolation. Server administrators are unconcerned with network switch configuration. Storage administrators dole out LUNs as they see fit. The network team controls all traffic. An inherent disregard of the bigger picture excludes efficient leveraging of infrastructure across the organization. Even architectures originally deployed with a holistic perspective tend to diverge as upgrades occur at a functional level.

The requirement to support a multitude of operating and management systems, protocols and equipment drives up costs while serial processing of technology requests leads to lengthy delays. The stovepipe model saving grace is that its shortcomings are often overshadowed by the even greater inefficiencies of a physical data center. It can, for example, take weeks of administrative time to provision new servers – making storage and network resource delays moot.

These same delays become excruciating against the backdrop of a virtualized data center where the server team can provision virtual machines in a matter of minutes. Virtualization also blurs the comfortable lines of functional segregation. Server administrators may need a separate VLAN to enable vMotion or extra SAN spindles to enhance performance. They configure virtual switches, argue about protocols and can take a LUN off-line by over-provisioning VM storage.

The silo model drawbacks are amplified as IT organizations deploy VMware vSphere, the Intel 5000 CPU and the Nexus 1000V virtual switch to obtain 100% data center virtualization. As physical servers become unavailable, IT can no longer circumvent bureaucracy on the fly.

Optimizing Virtual Machine Hosting with UCS

Just categorizing UCS is generating a lot of controversy. Last week HP called it a "giant switch". Howard Marks of Byte and Switch recently called it, "next year's servers…shipping this year". According to ComputerWorld, a Dell executive referred to it as a "one-size-fits-all blade server".

UCS can only be satisfactorily defined within the context of the virtualized data center that it was designed to optimize. It is a specially built hosting platform with innovative advances in both performance and management.

Cisco's patented technology delivers far more memory at lower prices than traditional servers. UCS's unified fabric capabilities resolve the cabling nightmare of hosting many virtual machines on a single server. Its Palo adapter supports up to 128 vNICs and vHBAs – providing select virtual machines with native storage and network characteristics.

UCS revolutionizes virtual infrastructure management by allowing the server administrators to dynamically assign resource pools created by the storage and network specialists to a physical blade. All teams share a common GUI and have visibility into what other groups are doing, yet their role specializations are preserved.

Transitioning to a Data Center Team

Resource pools are enabled through comprehensive service profiles that define UCS blades through software. Initial collaboration of the functional specialists is not only required for their creation, but for ongoing troubleshooting. Zoning and vSAN modifications, for example, should be communicated to the server and network teams. Changing a vHBA should be timed so as to not reset a production server outside of a maintenance window. The server team needs to monitor space utilization and performance of their systems to ensure LUNs are "right sized" by the storage team.

Specialists will likely come under pressure to, at minimum, pick up a strong secondary skill set. The server team needs some understanding of IP routing. The network group should know how to do minor server troubleshooting and basic reinstalls. Storage specialists need familiarity with proper VMFS sizing and best practices. All three groups should understand the implications of how security affects their domains.

The challenge and opportunity for IT management is to institutionalize regular communication, drive appropriate issues to a big-picture decision-making level and utilize increased accountability to promote cross-training. A good first step for many larger organizations is to restructure the storage, network and server groups to directly report to the same person. Creating a data center team will help enable a more efficient operation that is better able to optimize the extensive UCS capabilities.

Chris Reed, Mondy Carpio and Steve Jones of INX were invaluable contributors to this article.

Is VMware More Like Novell or Oracle?

David Cappuccio, in his blog for Gartner, Just a Thought; Will VMware become the next Novell?, draws several disconcerting parallels between VMware's current virtualization industry dominance and a similar position enjoyed by Novell in the early 1990's. Cappuccio wonders if VMware, like Novell, is going to get trounced by Microsoft.

The $60 billion Redmond behemoth is one of the greatest companies of all time and has displaced many industry leaders as viable players. Though it would be foolish to rule Microsoft out of any fight, differences in today's technologies and between the VMware and Novell organizations mean that history is not doomed to repeat.

Virtualization as a Data Center/Cloud Computing Platform

 

Microsoft touts Hyper-V as an operating system feature and arguably views virtualization as a point solution. VMware, on the other hand, is all about virtualization as a platform. This broader perspective has enabled VMware to dominate the market with exceptionally reliable, robust and versatile solutions.

 

While perhaps taking some liberties with terminology, VMware’s declaration of vSphere as a cloud operating system does emphasize the importance of virtualization in enabling a shared resource-on-demand cloud model. As a data center approaches 100% virtualization, embracing cloud computing becomes easier. VMware’s vSphere delivers not only the performance required for 100% virtualization, but also the crucial storage, network, security and management components.

 

Data center virtualization has an exceptional and easily measurable ROI. Achieving the full savings and other benefits, though, requires a homogenized virtualization platform. IT staffs will likely be reluctant to permit pockets of Hyper-V, even if only for test/dev or non-mission critical servers, making it more difficult for Microsoft to establish beachheads against VMware as it did against Novell with NT.

 

VMware’s tight partnership with Cisco adds a further enterprise credibility that Novell lacked. Cisco’s Nexus 1000V switch is practically a given for any Cisco shop wanting to consolidate virtual and physical network management, and Cisco’s UCS optimizes the effectiveness and efficiency of virtual infrastructures. The 1000V only works with vSphere while UCS is tightly integrated with the VMware platform.

 

The Difference between VMware and Novell 

 

In 1994, I ran the 2^nd largest Novell dealer in the San Francisco Bay area. Steve Ballmer invited me and a handful of other Novell partners to lunch in order to hear his pitch on why we should sell NT 3.5. I was particularly impressed by his response to a question about the impending Novell-Word Perfect merger. “We think it’s great!” he boomed. “WordPerfect has these huge budgets. Novell has tiny little budgets. The only thing those companies have in common is that they’re both headquartered in Provo, Utah. It will never work!”

 

Ballmer’s prescience quickly became apparent as Novell sold WordPerfect to Corel two years later. The acquisition was a casualty of Novell CEO Ray Noorda’s resolve to beat Microsoft at any cost – including purchases of Quatro Pro, DR DOS and even ATT UNIX. But by neglecting its core networking competency, Novell opened the door wide for NT to gain a foothold. 

 

Unlike Noorda’s Novell, VMware’s CEO, COO and Executive VP all come from high-level positions at Microsoft. They understand the Microsoft culture, strengths and threat. VMware is a company with a laser beam focus on virtualization that is undistracted by a personal vendetta.

 

Lack of Microsoft Invincibility

 

At the Microsoft MVP Meetings in the early 2000’s, Steve Ballmer commonly dismissed Oracle as a competitive threat, stating that SQL Server had already won. Yet Oracle now sits as one of the three largest software companies in the world with over $22 billion in revenues. Battling Oracle along with other exceptional competitors such as Google and Nintendo will prevent Microsoft from focusing its vast resources only on virtualization.

 

The ongoing improvement of Hyper-V and other hypervisors make a sustained 90% VMware market share unlikely. But VMware’s industry leadership need not diminish. The key will be its ability to continue innovating with virtualization solutions that increasingly obsolete the physical alternatives. As VMware drives the virtualization industry expansion, it will benefit in an absolute sense with continued rapid growth.

 

**** 

This article also available on www.dabcc.com.  Named by VMTN as one of the top 5 weekly blog posts.

http://blogs.vmware.com/vmtn/2009/07/top-5-planet-v12n-blog-posts-week-27.html

Will Cloud Computing See an ASP-like Burst?

The May announcement of vSphere’s debut elevated its status from a mere Virtual Data Center OS to its lofty designation as a Cloud Operating System. One can hardly fault VMware, though, for joining the marketing fray. Even Oracle’s Larry Ellison succumbed as he acknowledged, “We'll make cloud computing announcements. I'm not going to fight this thing. But I don't understand what we would do differently in the light of cloud."

Nearly every manufacturing and consulting firm in IT has seemingly hitched its wagon to the cloud as the hype has grown toward epic proportions. Just try to obtain a unique URL with anything cloudlike: “Versacloud.com” – Ha! “Silverlining.com” – Please. “Muteablecloud.com” – Forget about it. The media and research firms are all abuzz as well. Gartner says that cloud computing will hit $150 billion by 2013.

But then, we’ve seen this before.

The ASP Industry Implosion

A decade ago, ASPs suddenly emerged as the inevitable alternative to more expensive, less predictable and less responsive internal data processing architectures. Gartner Group boldly predicted that ASP revenues would jump from $900 million in 1998 to $23 billion by 2003. The research firm gushed, “Despite market debate, the shift toward the ASP model is inevitable…we believe this model will survive and flourish, simply because the benefits are indisputable.” Gartner was far from alone in its optimistic predictions. IDC projected a $24 billion ASP industry by 2005 while Zona Research estimated it could reach $48 billion by 2003. 

For fiscal year 1999, the top five publicly traded ASPs had combined losses of $485 million versus sales of only $464 million. Despite these horrific numbers, their total market cap reached an astounding $23.6 billion by March of 2000 (Table 1). Clearly, Wall Street shared the research organizations’ exuberance for the ASP model.

Unfortunately, someone forgot to tell the customers. While total fiscal year 2000 sales of the top five publicly traded ASP firms climbed slightly to $547 million (most of which was not due to hosting), total losses soared to nearly $1 billion. Not surprisingly, the subsequent burst of the ASP bubble was far more dramatic than the dot com collapse. By 2005 three had gone bankrupt, and the remaining two had combined sales of $135 million and a total market cap of $111 million – a 99.5% decline in total valuation.  

ASP	12 Month Revenues	12 Month Earnings	March 2000 Market Cap
FutureLink	$  99.9	-$95.2	$  2,147
Breakaway	$  98.6	-$43.5	$  3,680
Interliant	$128.8	-$125.1	$  2,294
NaviSite	$  49.8	-$  58	$  8,790
USInternetworking	$  86.9	-$163.6	$  6,686
Total	$464.0	-$485.4	$23,567

  Table 1: Top 5 publicly traded ASP financial metrics as of March, 2000

Cloudy Forecast

Cloud computing may be all the rage in the media, but market valuations of cloud providers are hardly in the stratosphere. VMware, for instance, recently paid $20 million for 5% of cloud computing provider, Terremark. This $400 million valuation is quite reasonable when considering Terremark’s rapidly increasing revenues and EBITDA which were $250 million and $51 million respectively for the last 12 months.

The overwhelming hype around cloud computing compels CIOs to evaluate it, if for no other reason than being able to demonstrate their diligence. But that doesn’t mean they’re clamoring to roll out cloud services. On the contrary, we’re seeing an unusual skepticism regarding claims of security and SLA capabilities. Organizations want to understand how they can optimally delineate internal versus external computing along with the trade-offs between cost, performance and risk.

Unlike the ASP industry which was doomed from the start by technology limitations, virtualization provides cloud computing with the foundation for technological success. The healthy caution with which organizations are approaching clouds, undoubtedly motivated at least in part by the difficult economic climate, will in the long run help shape a more practical and resilient cloud computing industry.

Why Cisco UCS is Kind of like the Kindle

I've been reading Cisco's new book on UCS titled: Project California: a Data Center Virtualization Server. The book is exceptional, but the title concerns me. The media and potential UCS customers already struggle with how to categorize UCS. By referring to it as a "server", authors Gai, Salli and Anderson may create further confusion. The term doesn't do justice to the new category that UCS creates – a compute platform unifying network, storage and server virtualization.

While obviously very different types of products, UCS and Amazon's electronic book reader nonetheless have quite a bit in common. The paperback sized Kindle 2 combines innovative display, wireless and low power technologies to provide an optimized hosting platform for electronic books. It eliminates the printing, distribution and storage costs inherent in the old-fashioned paper versions. UCS similarly incorporates innovative technologies in memory, network I/O and unified fabric management to optimize the virtual environment. It facilitates a pervasive operation of agile virtual machines that are much less expensive to deploy and manage than physical servers.

Kindle 2 is still rudimentary in some respects such as lack of a touch screen, but it makes so much sense that you know the traditional printed book market is in for long-term trouble. UCS similarly is going to redefine the very concept of "servers" which will increasingly become associated with virtual machines residing on a unified hosting platform. And the desktop market is not far behind. The "desktop" is rapidly being redefined as a virtual machine residing in the data center that is accessed with a PC, laptop, Mac, thin-client or zero-client device.

Hopefully I will be able to purchase the next edition of the UCS book for my Kindle – the printed version is so passé. I have to highlight sections using a marker and look up words with a dictionary. I even had to make my own bookmark from a magazine flyer since I threw away all my paper versions months ago.

Don’t Believe Any Numbers You Don’t Make up Yourself

Microsoft, in emphasizing the value of Windows Server, has stated that open-source Linux is "free like a puppy". It now argues the other side against VMware by constantly equating the "free" licensing cost of Hyper-V with value. The comparison fails to consider how the respective platforms stack up in their ability to assist organizations in achieving their ultimate virtualization objectives.

Where's the Beef?

McDonald's founder Ray Kroc liked to claim, "We take the hamburger more seriously than anyone else". VMware could say the same thing about virtualization. Every page of its Web site educates viewers on different aspects of virtualization, but the term is not even mentioned on Microsoft's home page. Navigating to Microsoft's virtualization section reveals the biggest emphasis to be on price comparisons between Microsoft and VMware. Videos, white papers, blogs, case studies, analyst quotes and even Microsoft's ROI calculator tool underline the cost competition with VMware.

VMware's suite of network, management and automation tools enable a virtualized data center, disaster recovery solution and desktops. Microsoft, alternatively, emphasizes its ability to address both physical and virtual infrastructure. While VMware stresses the superior performance, security and reliability of its hypervisor architecture, Microsoft promotes as advantageous that Hyper-V is incorporated as a feature of the operating system.

Unlike an operating system running an individual application on an isolated server, virtualization integrates formerly disparate silos of compute, storage and network resources. Virtualizing a production environment results in zero tolerance for issues in performance, reliability or security that now affect the entire IT operation. While the many unique capabilities of vSphere such as the distributed virtual network switch, storage live migration and integrated VM level fault tolerance may be argued to be essential for a virtualized data center, nothing is more important than reliability. Duncan Epping of yellow-bricks.com fame gave a succinct response when asked why he used VMware: "Because it's rock solid!"

The Microsoft ROI Calculator

Microsoft's Web site audaciously claims that Hyper-V is one-third the cost of VMware. As Table 1 shows, the cost of Windows licensing is factored in when purchasing VMware although this is only a requirement when running Windows VMs. It also disregards the obvious deployment of Windows Server Datacenter Edition which allows unlimited Windows Server instances on an ESX/vSphere host. Microsoft does concede in its "one-third the price" video that the Hyper-V to VMware cost ratio increases to 50 per cent if using Datacenter Edition.

Table 1: From Microsoft Web site showing Hyper-V 1/3^rd the cost of ESX (ESXi is a typo)

Microsoft also takes the very unusual approach of using its Virtualization ROI Calculator to compare ROI results with ESX. While an ROI metric can be useful in evaluating the economic impact of different platform investments, all of the relevant variables must be factored for each scenario. The Microsoft calculations erroneously treat Hyper-V and ESX exactly the same.

I ran a virtualization scenario on both the Microsoft and VMware calculators for comparison purposes. Both tools have a similar look and feel which is not surprising since they were created by Alinean, and both show a very similar 3-year net cost reduction from the physical scenario. I used the following parameters, but otherwise Microsoft default figures wherever possible:

• Analysis period: 3 years
• 350 total physical servers: 100 2CPU/1 core, 200 2CPU/2 core, 50 4CPU/2 core
• Virtualization hosts: 2 CPU / 4 core
• Users: 5,000 (the Microsoft tool requires an input for # of users in order to work)
• Cost of capital: 0 (for simplification purposes)
• Cost variables to consider: Existing servers only
• Virtualization: 100% (this percentage is fixed for both tools)
• VMs/host: VMware uses a ratio of 8.31:1 (consolidating to 42 hosts), while Microsoft uses a ratio of 5:1, consolidating to 70 hosts (for both Hyper-V and ESX).

The output from the Microsoft ROI tool, shown in Table 2, has a 3-year net cost savings for both the Hyper-V and vSphere scenarios of $2,083,449 – but an initial Hyper-V 3-year investment of $399,801 vs. $1,448,561 for vSphere. Table 3 shows a breakdown of the investment components for each solution.

Table 2: Output from Microsoft ROI Tool with 350 servers, 5,000 users.

Analyzing the Microsoft Calculator Investment Figures

INVESTMENT
		Hyper-V	vSphere
Windows Server Ent. Licensing		$ 88,113	$    88,113
CALs		$   -	$    78,400
ESX Licenses		$   -	$  402,500
vCenter		$   -	$   5,000
Backup Software		$   -	$  287,000
Storage		$ 73,500	$    73,500
Professional Services/Installation		$  238,188	$ 238,188
Training		$   -	$    10,000
Total Initial Investment		$  399,801	$ 1,182,701
Total 3-Year Investment		$  399,801	$ 1,448,561
ROI (over 3 years)		521%	71%

 Table 3: Investment Summary from the Microsoft ROI Tool

Windows Server Enterprise Licensing

The Microsoft tool assumes that all hosts run Windows guest OS's, and then calculates Windows Server Enterprise rather than Datacenter because of an assumed server consolidation ratio of 5:1 (which makes Enterprise less expensive). If the ratio is larger than 5:1, it quickly becomes advantageous to utilize Datacenter Edition.

Although the Microsoft tool assumes identical hosting ratios between Hyper-V and ESX, evidence suggests otherwise. Evaluations by our engineers indicate a ratio of about 2:1 in VMs per host under ESX versus Hyper-V. A quick Google search shows this ratio or higher to be commonplace. A ratio ranging between 1.5:1 and 2:1 was the conclusion of a VMware funded study by the Taneja Group. The 2:1 ratio is supported by comparing the internal consolidation ratios achieved by Microsoft as outlined in its January 2009 Technical Case Study as opposed to those achieved by VMware as presented in its January 2009 Webcast by CIO Tayloe Stansbury. A ratio of 1.67:1 is obtained by using the respective consolidation ratios from the VMware and Microsoft calculator tools.

A larger hosting density for VMware significantly lowers both the investment and three-year operational costs by reducing the number of virtualization hosts and their associated expenses. It also reduces licensing requirements for Microsoft products such as Windows Server Data Center Edition and SQL Server Enterprise.

CALs

$80,000 worth of CALs are applied to the VMware investment cost, but not to Hyper-V. I can find no reason for ascribing CAL costs to the VMware scenario, and assume it is a mistake in Alinean's methodology.

ESX Licenses

The "initial" cost of ESX licensing includes three years of annual subscription and support (SNS) at full list, although VMware offers a significant discount for purchasing SNS up-front. A greater hosting density would result in a decreased cost for both licensing and SNS.

Backup Software

In Microsoft's view, each ESX host requires an additional $4,100 in management/backup software to put it at parity with Hyper-V. Although the aggregate cost of $287,000 equates to 20% of the total VMware investment, I was not able to find an explanation for it.

Storage

The cost of storage appears to be significantly understated for typical virtualization situations. The VMware calculator tool reflects an estimated storage cost of $432,000 for the same scenario. VMware says that vSphere capabilities such as thin provisioning now reduce storage costs by over 50%, a claim appearing to be corroborated by testing results beginning to emerge from the field. For an organization with significant storage requirements, vSphere may have a substantial storage cost advantage over Hyper-V.

Professional Services/Installation

Microsoft likes to tout that Hyper-V is based on familiar Windows interfaces and works with well-known Windows-based technologies. Most midsized and larger organizations already utilize VMware meaning that IT is familiar with its technology and management tools. As an organization climbs closer to 100% virtualization, it is able to leverage its staff expertise to produce still greater gains with minimal further investment in training.

Training

The Microsoft tool applies a cost of $10,000 for training to the VMware scenario but not to the Microsoft scenario (though a note confirms it should be applied to both). This appears to be another Alinean error.

Return on Investment

The cost of the respective investment for both the vSphere and Hyper-V scenarios is subtracted from the savings to result in the net savings ($2,083,449 in Table 3), but then the cost for the Hyper-V investment is subtracted again from both numbers when calculating the ROI. This is clearly an Alinean computational error.

100% Data Center Virtualization

Up until now, most organizations have generally been reluctant to virtualize the majority of their production servers. My soon-to-be-published article, Perfect Storm for a 100% Virtualized Internal Cloud, describes how vSphere 4 provides the performance, fault tolerance, dynamic resource allocation, networking capabilities, security, management and API components necessary to successfully obtain 100% virtualization. The greater the percentage of data center virtualization, the more an organization is able to save and consequently leverage its virtualization investment.

VMware says its internal IT will be 100% virtualized by the end of the month in contrast to Microsoft which says it is shooting for an internal virtualization ratio of 50%. Although Hyper-V R2 will resolve some of the challenges Microsoft's IT department faces today, it will still be hampered by limitations such as an inability to apply network and security policies that are live migration aware. Even if a Hyper-V organization could virtualize 80% of its servers as opposed to 100% for vSphere, it would still suffer a markedly higher operating cost over the analysis period. The number of physical servers in a Hyper-V scenario is increased still further if an organization uses Linux other than SUSE.

Does ROI Matter?

Virtualization is a rare IT technology that provides a "hard" ROI without quantifying intangibles such as lowered risk and increased agility, or even IT staff benefits. Consolidating servers by 90% or more slashes associated costs such as power, cooling, cabling, power whips, generator and UPS slices, rack space, distributed network switch ports, core network switch ports, SAN ports, HBA ports and maintenance contracts. Over a period of a few years, these cost savings can easily amount to hundreds of thousands of dollars for organizations with around 25 servers or more, ranging up to millions or tens of millions for larger organizations.

Running varying virtualization scenarios using my own ROI tools all show vSphere as the superior economic choice. But as the title of this blog says, you shouldn't believe any numbers that you don't make up yourself. I encourage IT leaders considering virtualization to not just rely on manufacturer calculator tools, but to take the time to carefully evaluate the economics of different virtualization solutions in their unique environments.

ROI can be very helpful in deciding whether or not to virtualize, but it is typically not a good metric for comparing different products. The huge economic and other benefits resulting from a successfully virtualized data center inevitably make the cost comparison between individual components irrelevant in comparison to their ability to facilitate the desired outcome.

------------------------------------

Follow-up:  Tom Pisello of Alinean wrote a thorough response to this blog.  Alinean seems to agree with most of my points, and is even "posting an update to proactively address" the issue of CALs (this section of their tool was defined by Microsoft).  Alinean apparently did include notes regarding the backup software (I was unable to find them when writing the blog, but will look again when I get a chance), and found no calculation error with the ROI metric (I went through this carefully before reporting it - but since the error was repeated for both Microsoft and VMware scenarios, it wasn't material enough to warrant me running through all the scenarios again to replicate the issue).

Overall, I thought Tom's response was great - nondefensive and clearly representative of a sincere intention to provide organizations with as accurate results as possible.

Open Letter to Steven Chan of Oracle

Steven Chan is a senior Director in the Oracle Applications Technology Integration group.

Steven,

Your April 10, 2009 blog discusses Oracle's support policies on virtualization products such as VMware and Citrix, referencing Metalink Note 794016.1, which said that Oracle E-business applications are "not explicitly certified, but supported" on "both software and hardware based" virtualization technologies. The link to the note in your blog, though, is no longer active. 

In his May 14th blog, EMC's Jeff Browning reveals why.  He says that as of May 8, Oracle changed its support policy to no longer cover software based virtualization techologies.  Given Oracle's pending acquisition of Virtual Iron, Jeff concludes, "It appears that Oracle is carefully and intentionally excluding VMware from its support statement".  Jeff poses some questions for Oracle including:

• Given that there is abundant evidence from extensive testing (including at EMC) that VMware virtualization of Oracle databases is both viable and compelling, what is your real issue here?

• ...when will you extend official best-efforts support to VMware which you have already done to other equivalent virtualization products?

Many of our clients run Oracle on VMware today, and we anticipate that number to rapidly escalate now that Oracle 11G and even large Oracle OLTB applications run faster as virtual machines on vSphere 4 than they do on physical servers.  The ability of vSphere's Fault Tolerance to enable real-time fail-over of Oracle applicatioins will be another compelling driver to virtualize Oracle on top of VMware.

You make a point of emphasizing your "Blogging Code of Ethics" which specifies that, "This blog's "first priority is the needs of the Oracle community."  Given that your April 10 Blog is no longer accurate, and on behalf of the 18,000 VMware partners, I'd appreciate you clarifying Oracle's support stance so that we in turn can report back to our enormous base of Oracle customers who deserve to know your intentions.

Respectfully,

Steve Kaplan

Why We Sold AccessFlow to INX

As the June 9th anniversary of our sale of AccessFlow to INX approaches, I've been reflecting about our decision and how things have turned out.  We were fortunate in that AccessFlow’s success as a leading VMware partner led to a fair amount of interest from potential acquirers.  In early in 2008 my partner, Gary Lamb, foresaw the coming convergence of VMware and Cisco in the data center, and felt it important that we be aligned with a strong networking organization.  We agreed to be acquired by INX last June because of the firm’s stellar reputation as one of the nation’s top Cisco partners.

In our acquisition discussions with INX’s executive team, we made it clear that we did not want to become the virtualization division of a networking/UC company.  We were assured that INX was making the investment to acquire AccessFlow because the company wanted to become a leader in the data center virtualization space.

INX has lived up to the promise.  The firm quickly became a VMware Premier Partner, acquired a seat on VMware’s Partner Advisory Council and began hiring great VCP engineers at its branches across the country.  And while it was initially a challenge to convey the virtualization messaging to 125 account managers steeped in “speeds and feeds”, the receptivity has been fantastic.  I have to confess that it is a bit intoxicating to be ushered into large Cisco accounts across the country in order to talk about our virtualization vision to a very attentive audience.

It turns out that the Cisco and VMware partnering went beyond what we expected.  Cisco’s VN-Link technology is incorporated in the vNetwork Distributed Switch of vSphere 4, while VMware technology enables the existence of the Nexus 1000v , and to a large extent, the market for UCS and the Nexus 5000 and 7000 family of switches.  Products from both organizations are, in our opinion, essential if the objective is 100% data center virtualization – something that Gary and I have felt should be the goal since we co-founded AccessFlow in May of 2005.

Last week INX achieved another milestone in data center virtualization by being one of the first few companies to receive the Advanced Technology Partner certification for Cisco’s Unified Computing System. http://finance.yahoo.com/news/INX-Achieves-ATP-for-Cisco-bw-15169875.html?.v=1

Microsoft blog on Hyper-V versus VMware pricing

http://blogs.technet.com/virtualization/archive/2009/05/06/microsoft-hyper-v-server-2008-r2-release-candidate-free-live-migration-ha-anyone.aspx

(I wrote the following response to Microsoft's blog, but it turns out MS doesn't post reader comments).

I have spent 20 years developing, writing and speaking about ROI as the appropriate tool for making intelligent IT platform decisions.  TCO is inadequate because it fails to consider the context of the overall savings realized.  Such is the case here.  When organizations successfully virtualize their data centers, they are typically able to save hundreds of thousands, millions or tens of millions (depending upon the size of the environment).  Moreover, because the bulk of virtualization cost is absorbed up-front in terms of hardware (servers/storage), consulting and training – the marginal ROI that results as an organization virtualizes closer to 100% of its data center is even higher.

The enormous savings at stake make it imperative that an organization considering virtualization should evaluate its overall projected savings/cost avoidance versus the overall costs to achieve those savings.  Data center virtualization typically enables a truly amazing reduction in costs relative to the investment, resulting in payback periods commonly less than 12 months.  And these savings/cost avoidances are not the reduced risk or increased agility type of ROI commonly touted.  While virtualization certainly includes these benefits, we’re talking about only very hard savings of hardware, data center space and power that the CFO will validate and sign off on.  When calculating ROIs for our clients, we don’t even bother to quantify the large reduction in IT staff time required http://www.dabcc.com/article.aspx?id=9425, because this figure is a little more fuzzy and can therefore lead to skepticism.  Besides, there is no need – the other savings are compelling enough on their own.

The fist step to a successful virtualization project is therefore to calculate an ROI for which we only need a general idea of the investment in order to derive an order of magnitude for the savings/investment ratio.  Assuming the numbers overwhelmingly justify an enterprise virtualization deployment, the optimal platform (HS/SW) for successfully enabling the virtualization project can then be evaluated.  While the comparative cost of any specific solution can be a factor, much more important is its ability to facilitate the overall goal of 100% virtualiztion with all the performance, reliability, scalability and manageability required. 

 

A lot has been written about the superior capabilities of ESX to any other virtualization platform in these areas including features such as Storage vMotion, DRS, Distributed Power Management, etc.  Even Redmond Magazine, “The Independent Voice of the Microsoft IT Community” gave its 2008 Editors Choice award for the most reliable IT technology to VMware ESX (the IBM mainframe came in #2).  Now vSphere takes these capabilities up a level with new features such as Fault Tolerance, vmSafe, storage thin provisioning and around 150 more.  And the enhanced performance capabilities of vSphere mean that even the largest workloads can now run as fast or faster as virtual machines.

 

One of the major inhibitors to organizations virtualizing their data centers today has been the lack of networking support within any virtualization hypervisor.  This becomes even more important in facilitating a seamless cloud computing strategy.  vSphere includes the vNetwork Distributed Switch enabling QOS and the application of security and network policies that follow virtual machines as they migrate across hosts.  By adding the Cisco Nexus 1000v (which only works with vSphere), network administrators can use the same Cisco command line tools in the same way they manage physical switches.

 

The competition to VMware vSphere isn’t Hyper-V or any other hypervisor, it’s inertia.  By not virtualizing, organizations continue to purchase servers, network and SAN ports, rack space, power and cooling, UPS and generator slices, cable runs, etc.  It makes far more sense for an organization to virtualize today with the best platform available and achieve what is likely to be an amazingly short payback on the investment while also reducing IT staffing requirements.

 

Since I've covered the ROI, I do want to bring up a TCO item you highlighted. You say that you leave the cost of the guest OS out of your calculations because, "That cost is the same whether you're running Hyper-V Server 2008 R2 or VMware so I didn't bother to include it".  This is silly - if we're running a Linux VM on vSphere, we obviously, as opposed to Hyper-V, do not require a copy of Windows Server.

 

Finally, even when evaluating Hyper-V and vSphere on a TCO comparative basis, we still need to consider the context of the overall cost requirements for each.  We estimate that Hyper-V will require twice the number of servers as vSphere to handle a similar number of virtual machines (Microsoft and VMware’s recent documents reflect this ratio today for the organizations’ internal virtualization projects).  This means a requirement not only for twice as many hosts, but also more rack space, network ports, SAN ports, maintenance contracts, generator and UPS slices, etc.  Additionally, Gartner says that it now costs more to power and cool a server than it costs to purchase the machine itself.  So when factoring in all of the additional costs that Hyper-V requires up front, it doesn’t even compare favorably on an isolated TCO basis.

 

Will the Threat of Swine Flu Boost VDI Sales?

Despite the extraordinary hype around VDI and the rapidly burgeoning VDI industry, it’s difficult to find sizeable enterprise deployments.  Part of the issue is that of the two real contenders in the market, VMware’s View 3 has only been shipping for 4 ½ months, while Citrix’s Xen Desktop represents a complete paradigm shift away from the 98.5% of its sales that are Microsoft Terminal Server based.  While it hopefully will pass without much more pain, swine flu unfortunately does have the potential to accelerate deployment of VDI.

Citrix got its big break in 1994 when Novell started referring its resellers to Citrix's OS/2-based remote access product, WinView.  As the Citrix technology matured, though, it was not positioned nor implemented as a widespread hosted desktop solution.  The limitations of Terminal Server made such an environment more confusing for users and harder to effectively administer.  The simplicity of VDI is appealing to users because their virtual desktops have the same look-and-feel as their normal PCs.  It appeals to IT administrators because it eliminates the management challenges of an enterprise Terminal Server-based deployment.   

As with all big IT changes, though, VDI is taking time to work its way into corporate consciousness as a viable alternative to traditional desktop computing.  Just the threat of pandemic disasters such as swine flu may give IT managers an incentive to start setting up virtual desktops for users as a preemptive strike.  VDI would let users work from home or other remote locations as easily as from headquarters.  As was the case with Citrix WinView many years ago, this could spark many more instances of VDI.  The advantage of the new technology, though, is that now IT managers are familiar with the concept and savings of a server virtual machine.  It won't take them long to realize that they can achieve much greater benefits by making their remote access desktop virtual machines the organizational computing standard from headquarters as well.

Check out my article on VDI coming imminently on searchvmware.com.

 

VMware has its Head in the Clouds, but its Feet are Planted Firmly on the Ground

Note:  This blog has been transformed to an article on DABCC.com. 

http://www.dabcc.com/article.aspx?id=10627.  DABCC is a leading virtualization Web site with huge traffic.  Doug is also a very sharp observer who is literally world renown.

 

Accent on Sales

Our VP of Sales says that I have the best job in the company – and he’s probably right.  I get to go around the U.S. accompanying our account managers on data center virtualization (DCV) sales calls.  Although I’ve primarily run companies over my long career in IT, the favorite part of my job has always been getting in front of customers.  It’s especially a kick now that I work for INX to be introduced as a DCV expert and be able to participate so fully in the most enjoyable part of the sales process.

The experience with INX has also expanded my perspective because I meet with so many different organizations across the nation rather than primarily in Northern California.  One of the interesting things I’ve observed is the power of accents.  I’ve often wondered why accents tend to hang around in many places despite the globalization of our economy, the transitory nature of our country’s inhabitants and the media bombardment of primarily accent-free messaging (at least accent-free to those of us from California).  The answer became evident to me during a recent trip to Boston.  One of the local INX DCV team members had not previously been introduced to most of the folks with whom we were meeting.  Nonetheless, when he addressed them in his heavy New England brogue, they noticeably relaxed and responded with an interactive demeanor borne from a sense of familiarity and comfort.

I concluded that accents are a means by which people subconsciously perpetuate a sense of shared geographic culture.  When our Boston team member began speaking, the customers knew he was one of them – and immediately embraced him as an insider.  This probably explains why it is particularly difficult for outsiders to sell into tight-knit enclaves such as New England.

Sandler Sales Training is a popular course which includes a methodology whereby a salesperson is supposed to mimic the speech patterns of the customer.  Proponents swear that it works.  If true – it may be partially due to the whole accent/culture mindset.  It is one of my goals to continually improve in sales, whether in Boston or Houston, so I need to take Sandler a step furthah and think about larnin’ some different U.S. dialects.  Talk to y’all later.

Microsoft Vs. VMware: Internal Virtualization Usage

Virtualization.info http://www.virtualization.info/2009/04/how-microsoft-and-vmware-use.html and vInternals http://vinternals.com/2009/04/microsoft-myths-and-realities/ provide excellent analyses on recent internal documents/presentations from Microsoft and VMware respectively on utilizing their virtualization products internally.  Microsoft's internal use of Hyper-V from a January 2009 Technical Case Study http://technet.microsoft.com/en-us/library/cc974012.aspx and VMware's internal use of ESX from a VMworld Europe 2009 presentation by CIO Taylor Stansbury http://sessions.vmworld.com/mgrCourse/launchCourse.cfm?mL_method=player.

To summarize, VMware uses server virtualization internally for 97% of its servers, and the remaining servers will be virtualized by the end of this quarter.  Microsoft, on the other hand, expects to have 50% of their server instances running on virtual machines by the end of Q2.  VMware is able to handle twice the number of VMs per 8 core/32 GB host as Microsoft (Microsoft actually uses a combination of 8 core and 12 core hosts, so the discrepancy is even greater).  VMware's deployment has no performance or stability issues and is both flexible and easy to manage.  Microsoft, on the other hand, has this to say about their deployment:

"Because of the required brief outage every time a virtual machine is moved from one host to another, Microsoft IT found that coordinating the server update processes with virtual machine owners was difficult.  Because one physical host could contain several virtual machines, Microsoft IT had to communicate with each of the virtual machine owners and coordinate host server maintenance with virtual machine maintenance."

Microsoft discusses its limitation of 23 LUNs because each LUN must be available to the Parent Partition" (Host OS). Windows does not deal with drives in terms of numbers, it deals in terms of letters (of which there are 26).  "A" and "B" are traditionally reserved for disk drives and, as such, cannot be assigned to a LUN.  "C" is the local hard drive and cannot be assigned to a LUN.  Therefore, Hyper-V can only address 23 LUNs.  Microsoft says it had to acquire new storage frames that supported more LUNs because they ran out with its 1 LUN: 1 VM strategy.  

Microsoft says it uses 802.1q trunking but not 802.3, and doesn't support NIC teaming.  This means that there is no network redundancy for the VMs in a Hyper-V environment.  This is a huge failing on the network side — and much more important than VLAN tagging they do support.  If a VM network port goes down, all of those VMs are offline until an administrator intervenes (Microsoft plans to fix this in the R2 release of Server 2008).  VMware, on the other hand, not only fully supports 802.3 and NIC teaming today, but with the vNetwork Distributed Switch (VDS) of the upcoming vSphere, will span many ESX hosts allowing quick scale up of networking capacity.  vNetwork also includes Private VLAN support, network VMotion, off-loaded network I/O and the ability for organizations such as Cisco to build VMotion-aware networking applications on top of it.  Cisco's Nexus 1000V virtual switch is expected to be announced on 4/21 in conjunction with vSphere.

When a VMware virtual machine wants to communicate with the network, it must be processed by the Guest OS and then by the Hypervisor.  Once the Hypervisor gets hold of it, the traffic gets pushed out to the network at large.  On Hyper-V, the traffic must be processed first by the Guest OS, then by the Hypervisor, and then by the Parent Partition (the Host OS).  The added layer (host OS) is the one that has traditionally been a bottleneck (which is one reason why ESX performs so much better than Hyper-V).

Here is a summary of the internal virtualization environments at VMware and Microsoft 

 

  Internal Virtualization Environment    VMware Microsoft

Current % of servers that run as virtual instances   97% < 50%

% of Servers that will be virtual instances by end of Q2    100%   50%

# of Production VMs per 8 core/32 GB host     10.4     5.7

High Availability enabled?    Yes No

Migration of live server instances without downtime?    Yes     No

Natively supports NIC teaming (network redundancy)      Yes No

Can add vSwitches without requiring host rebooting  Yes    No

Efficient communication of VMs with the network    Yes    No

LUN limitation    255    23

# of VMs managed per administrator    145 ?

Thanks to Jason Coleman and Steve Jones of INX who explained the networking and storage issues for me (though any mistakes in transcribing their remarks are solely the author's).

 

 

Presentation on Hyper-V vs VMware Sponsored by Microsoft and Dell

Hyper-V vs. Market Leaders http://www.bitpipe.com/detail/RES/1237562441_234.html

IT expert Nelson Ruest has impecable credentials as a well-published author.  Then again, I've co-authored many Osborne/McGraw-Hill technical books, and you wouldn't want me anywhere near your data center.   But even I understand enough to see that this is a pretty weak attempt to show Microsoft Hyper-V as a legitimate contender to VMware.  Take for instance, the only criticism of VMware from the presentation:

“Hyper-V is based on Windows Server 2008 and because of that you can run these machines on pretty well any hardware configuration, any hardware configuration that is designed to support Windows.  So, that means that you have hundreds and hundreds, if not thousands, of configurations that you can run Hyper-V on  whereas because ESX Server, VMware is more…a more limited edition…a more limited product because of that.  You can only run VMware on maybe dozens or fewer number of server configurations than you can run Windows.”
 
Is this really relevant to anything?  How many ESX host options does any organization possibly require?   Furthermore, new VM hosting platforms such as Cisco's UCS will start to have a major impact in data center virtualization, and UCS is optimized for VMware.
 
Check out these links which do a great job of dispelling Microsoft FUD:
 
http://itknowledgeexchange.techtarget.com/server-virtualization/vmware-fires-back-against-microsofts-mythbusters/
http://vinternals.com/2009/04/microsoft-myths-and-realities/
http://itknowledgeexchange.techtarget.com/server-virtualization/microsofts-embarassing-hyper-v-vs-vmware-video/
http://vteardown.com/2009/04/06/how-stable-is-hyper-v/
http://www.vcritical.com/2009/04/would-you-buy-a-hypervisor-from-these-guys/
 

 

 

 

Burton's Report: The Dark Side of Virtualization (4/6/2009)

http://www.burtongroup.com/Research/PublicDocument.aspx?cid=1464

Unfortunately, unless you're a Burton subscriber, you can't read their document.  Nonetheless, I'll go ahead and address the issues it raises.  Burton has a lot of clout in the industry, and although I disagree with some of the specific content of this paper, I recommend subscribing to their service for those who value generally well researched reports.

At a high level, I agree that virtualization has a dark side.  The dark side, though, has little to do with the purported issues presented in the Burton report.  Rather, it’s an implemenation done without the financial analysis, assessment, planning, proof-of-concept testing, piloting, beta deployment, project management and the carefully executed roll-out that should be a prerequesite to any successful enterprise deployment.  This tends to frequently limit the expansion of VMware deployments that were implemented originally as point solutions, and then began expanding beyond the original intention.  The organization consequently misses out on tremendous cost savings and other advantages while having to manage both a physical and virtual environment.

 

I think that The Burton Group knows this – they clearly reach during the document for any “dark side” bits that they can drum up in an attempt to write an interesting paper.  Their conclusioin, though, states that “Virtualization is an overwhelming success story in most IT shops and delivers on promised ROI.”  I agree with everything the conclusion says, including “awareness and taking action to address these new challenges”.  The interesting thing, though, is that they just tip-toed around the only true challenge of successful virtualization (other than lack of treating it as an enterprise deployment as mentioned above), which is the network. 

 

While the VMware virtual switch has been a remarkable way to achieve much faster inter-virtual machine communication on an ESX host, it limits the enforcement of security best practices which in turn impacts the environments that can be virtualized.  The VMware virtual switch is rudimentary; it is a basic layer 2 switch that doesn’t route, does not support QOS and provides no visibility to individual virtual machines.  Virtual machines running on an ESX host communicate with each other without going through the physical network –administrators cannot apply traditional security practices such as access control lists and private VLANs.  Administrators also have no way of easily isolating traffic to an individual virtual machine, and even if they do, a VMotion or DRS instance leaves them in the dark once again.  Managing virtual switches is painful because each one must be administered on an individual basis.  Since the server administrator is doing the configuration rather than the network team, misconfigurations and inconsistent policies often result.

 

Fortunately, the Virtual Distributed Switch (VDS) of vSphere addresses these issues.  And Cisco shops will be able to purchase the Cisco Nexus 1000V which makes managing the vSwitch just part of the overall Cisco network management scheme.

Let’s look at some of Burton’s “dark side” points:

 

Organizational Management

 

Burton claims that the difficulty of managing storage and network along with servers can cause a challenge.  Again, if a virtual infrastructure is put in without planning, then this could be a true statement.  Incorporating planning, however, enables a simpler environment than physical – even with the existing networking deficiencies.  Here’s an article I recently wrote about this topic that should shed some light  http://www.dabcc.com/channel.aspx?id=232.  Again, with proper planning and training, IT staffs inevitably really appreciate all of the advantages of a virtualized data center.  If you ever go to VMware’s VMworld, it’s like a 12,000 person love fest.

 

 Systems Management and Monitoring

While the performance and management tools are certainly evolving, again I’d argue that they are already superior to anything you can do in a physical data center given proper assessment/planning is required.  A good example of this took place with one of my clients a few years ago – when the technology was much less advanced than it was today.  An IT manager for a bay area bank told me that he liked everything that I told him about virtualization, but he worried that his security guy would never allow it.  I asked him, “How many servers do you have again?”  He replied, “Between 250 – 300”.  “But you don’t know exactly?” I asked.  “Not exactly,”  he said.  “If I asked your security guy how many servers you have, would he know?”  He replied, “IF I don’t know, then he certainly won’t know.”  I said, “So how is a virtual environment where you can view every server at any time, know the individual and aggregate resource consumption and know who was on what server at what time than in your existing environment where you don’t even know how many servers you have?”

 

Security

 

Burton makes a big deal about VMware having 22 vulnerabilities.  Every OS has vulnerabilities.  Even Cisco IOS has vulnerabilities.  The big question is whether there is any Exploit code available that goes along with the vulnerabilities.  As far as I know, the answer is no.  I’ve yet to hear of a single case of a VMware bypervisor-based security breach.  VMware even provided the NSA with their source code several years ago, and the NSA couldn’t hack between virtual machines.  The VMware hypervisor is extremely thin (probably over 1,000 times fewer lines of code of Windows Server), and therefore very difficult to exploit.  Burton Group again hedges here by saying that “Virtualization increases risk by adding complexity, but can be used to reduce risk by providing another means of separation.”  As my banking example above shows, when you virtualize, it is intuitive that fewer control points and much more granular monitoring and tracking capabilities in turn reduces security risk.

 

Burton says that, “customers will have to spend a lot on upgrades from the vendors for so-called ‘virtual appliances’.  First of all, virtual appliances are not by any means a mandatory part of a VMware deployment.  They are terrific, low-cost VMs that are generally superior to the physical versions – but are certainly not mandatory.  They also bring up the network security vulnerability, but as I mention above, that is addressed fully in vSphere and can be handled today through other means with proper planning.

 

Compliance and Audit

Clearly regulatory compliance is simplified in a VMware environment.  We were seeing this four years ago with the very heavily regulated pharmaceutical industry.  In that industry, products can be life or death.  When you get a prescription filled, it better be exactly the same as the product you received two years ago.  Regulation trumps all, and the FDA requires rigorous “validation” of all servers.  But since VMware virtual machines are provisioned from software and are thus 100% identical, pharmaceutical industry companies have long found it easier to receive and maintain FDA validation of virtual machines than their physical counterparts.

 

Vendor Licensing and Support

 

Burton first complains about the MS policy prohibiting movement of server OS licenses between hosts more than once every 90 days, and then goes ahead and gives the obvious resolution which is to simply purchase Windows Server Data Center Edition for each ESX host.  What they don’t discuss is just how economically advantageous this can be. 

 

Suppose, for instance, that an organization has 100 copies of Windows Server Enterprise.  We’re going toconsolidate those onto three Dell R910s with four 6-core CPUs each and 256 GB RAM.  Assuming that Windows Server Enterprise and Windows Server Data Center Edition both go for around $2,000 street price (they both list for around $3,700), then in the physical realm this organization has to spend $200,000 on Windows Server license (Windows Server Enterprise is licensed by physical server).  In a virtual environment, however, they simply license Windows Server Data Center Edition for the underlying CPUs of the ESX hosts, and can then run unlimited instances of Windows Server.  So the cost becomes $2,000 X 4 X 3 = $24,000.  The organization therefore saves $186,000 every time they renew their Microsoft Server licensing.  Other large licensing advantages can be obtained with SQL Server.  And with vSphere, there will be no reason to purchase Microsoft Clustering software any longer – thereby still saving more money. 

 

Regarding Oracle licensing in a VMware environment, well – I don’t understand Oracle’s licensing model (but then, I’m not sure anyone else does either).

 

Storage, Networking and Architectural Considerations

 

Again, proper planning will address these issues.  And VSphere, with vStorage and the Virtual Distributed Switch, takes storage and network management to a level beyond anything possible in a physical environment.  SRM already enables storage management for disaster recovery beyond anything possible in a physical DR scenario.

 

Burton says they advocate going with low server consolidation densities on blades so that a server failure doesn’t impact many servers.  This is baffling to me as, 1)  Servers rarely fail and 2) If a server does fail, the VM is off-line for only an average of seconds while it reboots – since it’s software, it reboots very fast.  Of course, with the Fault Tolerance feature of vSphere, there will be no down-time at all so Burton’s argument is totally moot.  We also disagree with their advocacy of blades for most situations as we wrote in an article last year, and our opinion is shared by most virtualization industry experts with whom I have spoken http://www.dabcc.com/article.aspx?id=9114.

 

Burton says that the ROI may not be there for smaller organizations.  We’ve found that organizations with as few as 7 servers can still achieve a nice ROI, even when purchasing a new SAN as part of the deal.  But, the key of course is to conduct the ROI analysis prior to virtualization in order to make sure.

 

Burton says that, “Virtualization sprawl has been a potential worry that has only been kept in check by the added management overhead that an IT shop has incurred to manage multiple virtual platforms.”  This is kind of silly.  Most physical data centers still have server sprawl despite the large expense and effort it takes to purchase, configure, rack, stacked, cable, connect, power, cool, protect, back-up and maintain physical servers.  Virtual machine sprawl can easily be contained either through policy or through automated workflow utilizing VMware’s Lifecycle Manager.

 

Recommendations

Burton says if you have an ROI of 18 months or less, you should virtualize (actually the terminology should be payback period as ROI is not a unit of time measurement, but we’ll let that slide).  I don’t think you can set a hard number in any case.  It depends on what else the organization can do with the money, and what other not quantified benefits result – examples being advantages such as no more server failure, no need for hardware-based maintenance windows, greatly enhanced disaster recovery, and on and on.  The rest of their recommendations pretty much imply that organizations should plan their deployments.  I agree.