The Console

Posted by Winston Bumpus
Director of Standards Architecture, VMware

With the launch of VMware vSphere™ 4 and the free VMware OVF Tool 1.0 this month, we have implemented support for the DMTF OVF 1.0 specification across a broad range of VMware products. The VMware vSphere 4 products, VMware vCenter™ 4 and VMware ESX 4, have built-in OVF support at both the API level and directly in the VMware vSphere Client. The stand-alone VMware OVF Tool 1.0 brings OVF 1.0 support to VMware Workstation, VMware Server 2.0, and earlier versions of VMware vCenter and ESX.

As part of the VMware vSphere launch, we are also making many of our products available as OVF packages, so they are readily available to deploy onto your data center infrastructure or your desktop. These products include the VMware vSphere Management Assistant (vMA), VMware Studio 1.0, the technology previews of VMware vCenter Mobile Access (vCMA), VMware vCenter Server 2.5 on Linux, and VMware vCenter Admin Portal.

VMware has been actively involved in the development of the OVF specification since the beginning (the initial draft specification was submitted by VMware, Dell, IBM, Microsoft and XenSource to DMTF in September 2007), and our first product to ship with OVF support was VMware vCenter 2.5 and VMware ESX 3.5 (February 2008). The OVF support in those products was based on the preliminary 0.9 specification and did not cover all aspects of the specification. Essentially, it was limited to import/export for single VM packages. With the release of VMware vSphere 4.0, we support the DMTF OVF 1.0 standard in full, which adds a significant set of new features:

• Support for importing and exporting multi-VM OVF packages (vApps)
• Support for the OVF environment and OVF properties (deployment time software customization)
• Support for the OVA format (distribute your packages as a single file)
• Support for advanced disk compression using delta-disk hierarchies (smaller packages)
• Support for automatic IP allocation and customizable URL links in the VMware vSphere client. (never need to go to the VM console anymore when installing a virtual appliance)
• Backwards compatible with OVF packages generated by earlier products (ready to use)

On the VMware vApp Developer blog, the engineering team behind OVF will start to dig into the details about all these new features and how they can be put to practical use - both by IT administrators and virtual appliance authors. So if you are wondering about the technical details and how to apply OVF in practice, that is a good place to go to learn more.

VMware is pleased to be a part of the newly announced DMTF Virtualization Management (VMAN) Forum. The group will promote interoperability between products that support the VMAN standards. The forum will develop and conduct programs to determine specification conformance in the areas of system virtualization management, virtual system management, and the open virtualization format.

We believe that OVF is an important standard to enable interoperability as well as increasing capabilities while greatly reducing the time for customers to deploy new applications and services.

[Update 2: Updated the blog link to point to the new VMware vApp Developer blog.]

[Update 1: Clarified first sentence: a technology preview of the VMware OVF Tool was released last year, but version 1.0 of the tool was released this month on May 21.]

Posted by Dan Chu
Vice President, Emerging Products and Markets

All clouds are not created equal.  Google recently posted a blog entitled “What we talk about when we talk about cloud computing” which outlines what Google perceives as advantages to its approach to cloud.  While VMware agrees that a major part of future enterprise architectures will reside in the cloud, we differ greatly on our approach.

To save everyone the time and energy, the summary of their post is essentially that Google uses cheap hardware that they expect to fail and smart software to build the equivalent of a giant computer, that Google AppEngine can deliver the cloud for traditional IT, and that the Google model can produce the fastest innovation for end customers. 

Google has a valid and interesting model but we are finding that it simply doesn’t work for the vast of majority of business IT.  Let’s address the issues individually:

1. Building the Giant Computer

Google says that they take “a large set of low cost commodity systems and [tie] them together into one large supercomputer” with their software.  Fundamentally, we agree with this approach as we have done exactly this for thousands of customers over the last 10 years.  Using virtualization, VMware’s solutions have become the de-facto standard for customers looking to improve the efficiency of their datacenters while saving costs.

Over 130,000 customers run VMware and over 55% of VMware datacenter customers have standardized on the VMware platform.  Companies like Lockheed Martin and GE are building their own internal clouds using VMware.

While we are aligned on the overall direction, the Google blog claims that their scale and approach of managing servers lends a key advantage.  If companies had unlimited resources and were able to build massive datacenters with all of these commodity servers, the Google model may be the way to go.  However, this isn’t the picture of most datacenters today.  What virtualization is able to provide is improved performance of applications, improved utilization of existing resources and nearly unlimited scalability. And VMware offers what many customers require—choice. In fact, more than 500 service providers including major global players like  AT&T, Savvis and Terremark offer VMware as their platform to deliver services.  In a recent survey of top global managed service providers, every one of the top ten ran on VMware.

The VMware vision is to enable customers to run the giant computer through software on top of standard hardware, to be able to choose seamlessly to either run in their internal cloud or in an external service provider cloud, and to provide connectivity and consistent manageability between the two.    

2. Leveraging an Enterprise-class Cloud

Google follows up by promoting its AppEngine stack as the way to deliver capacity and scaling for applications and databases, “to deliver the set of scalable services that customers would otherwise have to maintain themselves in a virtualization model.”

This sounds good, until you run into the issue of trying to run your core applications on AppEngine.  Customers are looking to match their IT platform to their business needs, not the inverse. The Google approach calls for a least common denominator set of non integrated cloud services that everyone squeezes into. Customers want the flexibility and breadth of solutions that exist today along with the efficiency of the cloud.   Customers are not about to re-write or modify their applications so that they can run in a specific cloud.  In particular, given current macro-economic circumstances, customers have a high priority for a cloud platform that can take their existing apps, and enable them to take advantage of the cloud.  

Finally, customers want advanced business continuity, availability, and management capabilities for production, enabled by such technologies as VMotion, High Availability, Fault Tolerance, Storage VMotion, and SRM.  Today customers are broadly using these capabilities and VMware’s service provider partners are also delivering these as a service.  For example, T-Systems has built a solution practice and large customer footprint for running SAP implementations in their VMware-based cloud.

The Google blog further suggests that “there is limited value to running an Exchange Server in a virtual machine” and that customers should just use Gmail.  Enterprises aren’t going to move off an enterprise class mail platform for a personal-use platform.  To take Exchange as an example, it represents the kind of business-critical core IT application that is what customers want running on VMware today.   It is consistently one of our top several workloads being run on VMware, and we have invested in a lot in ongoing performance work with Exchange. In fact, IBM has demonstrated industry-leading capacity of Exchange mailboxes where customers can scale Exchange greater horizontally with VMware than natively on regular hardware.  

3. Innovating with Cloud

The Google blog closes by asserting that “IT systems are typically slow to evolve,” and that Google is much faster to innovate.   This supposition is mostly focused on Google Apps and its pace for new feature rollout.  This is fine for customers who are looking for exactly the features that Google happens to be working on, but for any other IT needs that a customer might have, the Google stack is a black box to the customer without the component architecture that lets many different types of partners integrate and contribute their new technologies.  In contrast, VMware embraces and extends the entire X86 ecosystem of thousands of ISV’s along with hundreds of service providers to bring the cloud to customers on their own terms. There is nothing like that for Google’s cloud.   

Summary

For customers looking to maintain the flexibility to move back and forth between the external cloud and internal IT, Google’s proprietary platform is like the “Hotel California or the roach motel” where your apps go in, but they never come out.  

VMware provides choice—allowing customers to put new and existing applications where they want, when they want.   So in a nutshell, if you’re a business of pretty much any size and want to consume cloud services, there are options that allow you more flexibility, reliability, compatibility and mobility. 

And if you’re already a VMware user, leverage what you know about the reliability and compatibility of VMware VMs to run the cloud internally or with cloud providers worldwide like SunGard, T-Systems, Tata Communications, and many more. With VMware, the cloud isn’t something “out there” as simply a destination to get to; it’s a new “state of IT” that can also come to you.

Dan Chu is vice president of Emerging Products and Markets for VMware.  He leads VMware’s efforts in the areas of small and medium business, virtual appliances, and cloud computing.

Posted by Winston Bumpus
Director of Standards Architecture, VMware

After last week’s exciting VMware announcements around the first cloud OS VMware vSphere 4, this week we have additional important news around the announcement of the Distributed Management Task Force (DMTF) Open Cloud Incubator. As President of the DMTF I see this as an important activity at this point in the evolution of this technology as it has begun to revolutionize IT infrastructure. This group plans to tackle several key issues as it relates to the interoperability of cloud management. 

It is probably a good idea at this point to describe what an incubator is and why it is different from a regular work group. I think the best way to describe this is to say that the incubator is really a place for ideas to be developed or incubate before specifications are created. A while back, the DMTF wanted to encourage groups who were in the early stages of developing technology to come to the DMTF and use the existing tools and policies for the rapid development. So to that end, the DMTF borrowed a model used by several member companies and many industry standards organization and created a new DMTF initiative called an incubator. The intent of an incubator is to develop recommendations and draft specifications which will then be taken through the traditional standardization process. The hope was to encourage organizations and members to bring their ideas, even in an early stage, to the DMTF.

The other difference between an incubator and a traditional work group is that it has a leadership board that is comprised of the key stakeholders in the creation of the work. This organization at launch has 13 leadership board companies including VMware, in addition to many other participating companies. We hope others will join the activity and other companies may be added to the leadership board over time. 

From the charter posted on the DMTF site, this incubator plans to:

• Enable the use of cloud computing within enterprises and improve the interoperability between cloud platforms via open cloud resource management standards.
• Increase awareness and support by management systems vendors that develop products to manage cloud resources.
• Enable cloud service portability.
• Provide management consistency cross cloud and enterprise platforms.

One of the important activities of this incubator will be to develop recommendations for enhancements and extensions to the Open Virtualization Format (OVF) that may be needed for cloud computing. There has been quite a bit of discussion from me and others on how this may be an important building block for cloud interoperability. The incubator will be a good place for the industry to have those discussions and make agreed upon recommendations. The other two important aspects of this incubator will be around cloud APIs, particularly as they relate to management as well as looking at security models and methods to support this level of cloud interoperability.

VMware is pleased to be a part of this activity and hopes to help enable new levels of interoperability and choice for our customers.

Posted by Winston Bumpus
Director of Standards Architecture, VMware

To continue my previous post, there is lots of work and discussion going on amongst various technologies vendors, service providers and customers around cloud computing and the need for industry standards. This week we see the announcement “SNIA Forms Cloud Storage Technical Work Group.” VMware is a supporter of this release and sees this as an important set of interfaces that need to be standardized  to enable interoperable cloud computing. This new group will be a technical working group (TWG, pronounced twig) of the Storage Networking Industry Association (SNIA). This announcement is a milestone as it is the launch of probably the first true standards activity for clouds and was announced at the Storage Networking World being held in Orlando, FL this Monday.

So what is cloud storage? According to the SNIA, Cloud Storage can be contrasted with SAN/NAS storage as both are “Storage Networking.” However, provisioning may be different and how you pay for it may be different. One other primary difference is that essential management tasks for storage resources are usually performed by the cloud operator and may not be done by the storage users themselves. Also the cloud has bandwidth, latency and locality issues that will be reflected in the access models that will need to be used. So standards around data path APIs as well as provisioning and metered will need to be defined to allow for interoperability between enterprises, private and public clouds.

This new work will focus on SNIA architectures and best practices related to cloud storage technology. SNIA has been around for years and is a well-respected trade association and advocate for the storage industry. This new work stems from both the need for such an association and from discussions that have been ongoing in several public meetings and community forums.   Much of the creation of this activity has been done in conference calls, and on the “Cloud Storage” Google group. Currently there are over 25 organizations that have signed up to support the activity. Support of these efforts means more than having your name on the press release. It means that you commit to the hard work it takes to develop, negotiate and implement standard interfaces. I don’t believe this work will need to start from scratch as there are several technologies and standards that have already been developed that can be leveraged for this work. Certainly the use of and extension of existing standards, such as SMI-S (Storage Management Initiative Specification), will be an important part of this work.

As I mentioned in my last post, cloud standards will be developed in various groups as it is broader in scope than a single organization can address. SNIA is one of the first to announce plans and form a working group to start this work. I expect other groups and organizations to do the same as it relates to their areas of expertise. In future posts I will be providing additional background on the other work going on in the various standards organizations as activities and specifications continue to roll out  in support of  open cloud computing.

Posted by Winston Bumpus
Director of Standards Architecture

There has been a lot of discussion in the press and at various events in the past months regarding Open Standards for Cloud Computing. As a VMware employee and with my role as president of the DMTF,  I have a unique view of these emerging standards efforts. VMware supports the idea of cloud standards and has already made good progress on some key pieces. As one of the original authors of the Open Virtualization Format (OVF) specification, VMware has already shown its leadership and support for working with the industry to drive interoperability standards. In fact VMware believes that interfaces should be open so customers can have choice and improved interoperability while service providers  can differentiate on functionality and performance of their services.

The Distributed Management Task Force (DMTF) just announced the release of OVF 1.0 as a DMTF standard and suggested that OVF just might be a key building block for cloud computing interoperability.  Other interfaces and protocols also need to be standardized, which is why VMware announced its intention to submit its key elements of the vCloud API to an existing standards organization for the basis of developing an industry standard at VMworld Europe in February. We believe that the combination of standard packaging and deployment format along with a standard open API will go a long way to enabling cloud market growth.

VMware agreed to be part of the Open Cloud Manifesto. This document and discussion, while providing a very minimal set of principles to agree upon, will form a basis for initial agreements as the standards for this new computing paradigm are developed. We don’t believe there will be a single standard or standards body that will standardize all aspects of cloud interoperability.

I had the privilege to participate in the Strategies and Technologies for Cloud Computing Interoperability
(SATCCI) held in Washington DC area on March 23 hosted by the Object Management Group (OMG). Representatives from several standards organizations and communities were present, including the Storage Networking Industry Association (SNIA) the Open Grid Forum (OGF), the Cloud Computing Interoperability Forum (CCIF) and others. We agreed that one organization could not develop all of the standards needed, but that the various standards organizations and communities should work together by communicating which parts of the stack we would each address.

I will provide more background on the work going on in the DMTF and other standards organizations to drive new levels of interoperability in Cloud Computing in my next post.

Posted by Réza Malekzadeh
Sr. Director, Product Marketing & Alliances

Looming on the horizon are the nimbus, cirrus, stratus and cumulus that threaten to deliver us cloud computing imminently.  Promising an end to most of the challenges and frustrations of IT systems as we know them, the concept of cloud computing is thundering through the business community to become one of the most talked about and revered subjects of the day.

Behind the hype seems to be a reality that, for once, the IT industry maybe onto something truly game changing that will not only radically cut costs, but also deliver a far better experience to the business or consumer user. 

The expectations are huge. banking analysts say that cloud computing will be a $160 billion market within the next five years, and every major IT company from Microsoft to Google, from IBM to Dell, is desperate to be the rainmaker.

The question that comes to mind though is not “what” cloud computing is, but rather “why”. If it is such a great idea then why has it taken until now for the gurus of technology to deliver it. 

The “what” question is just too easy; imagine a world where you could walk up to any computer, anywhere in the world and instantly access all your data and applications just as you left them last time you logged on – and somewhere, up in the clouds, a huge IT infrastructure was whirring and churning to deliver the IT services to you.  Basically, think of the ease of getting electricity from a socket in your home that somehow connects to a generating station and you start to get the idea.

Why has it taken so long? Go back far enough in time and IT professionals always thought that computing would be delivered from the cloud and that the personal computer was nothing more than an aberration.  Early mainframes where constructed to deliver IT services down wires to dumb terminals that could do no more than display text on a screen and take back digits typed into a keyboard.    

These mainframes could handle hundreds or even thousands of users and if they had carried on evolving then, we would probably have had cloud computing in 1988 - rather than 30 years later.   

In fact, Thomas J. Watson, founder of IBM, is supposed to have remarked that “there is a world market for about five computers”.  He didn’t mean that these new fangled devices would never catch on (as Lloyd George unfortunately said about TV) but that his vision was of a few massive number crunching mainframes in the sky that could deliver their computational power to the users remotely.

What was not understood though, was the challenges that cloud computing would have to overcome.  And this is where the answer to the “why now” question lies.

To deliver cloud computing requires five critical components: the scalability of the infrastructure to meet users’ needs; the resilience to accommodate the unexpected; the network to distribute the applications; and the ability to deliver an acceptable experience to the user at a reasonable cost.

When it came to scalability the reality was that you built or rebuilt your datacenters once every five years to fit an estimated workload or users and traffic.  The concept of “dynamic” or “on demand” capacity existed only as a concept.  But something fundamental changed at the start of the 21st Century, when server virtualisation suddenly arrived on the scene, as a result of innovations led by VMware.

Where previously you had attached a given application to a server only to see users slow to a deathly halt during periods of peak usage, now you could now decide to vary the server capacity or resources available to a virtualised application and so scale it up or down according to demand.  This was freedom for the CIO and MIS staff as they suddenly could adapt their business to the needs of the user community.  It wasn’t cloud computing yet, but maybe the forerunner.

Resilience was probably the biggest killer of the original IT model and gave rise to the PC almost by itself.  Despite cloud computing being the ideal solution for IT architecture, the repeated and sustained or catastrophic breakdowns of mainframes led users to revolt against the tyranny of the IT director.  The phrase that sends shudders though the souls of many middle-aged ex-programmers is “unscheduled outage” as hours of work would be lost to some minor bug on a given server.  Evolution eventually kicked in with the concept of transferable workloads made possible by innovations such as VMware’s VMotion – a technology which can take a running application from a problematic server to another server with no interruption. 

Bizarrely the network was probably the least of the problems.  Arpanet, the forerunner of the Internet, was up and running in the 80s and although designated for military use quickly proved itself within the academic community.  But at an original 50 Kbps, compared to today’s multi-Megabit throughputs, there is no doubt that broadband has transformed the landscape for cloud computing.

Probably the most emotive issue in IT is the end-user experience. Grown men have cried at the prospect of rebooting Windows Vista and previous experiences of cloud computing were little different. We expect and have a right to an IT experience that delivers the goods.  An ATM machine, a great example of existing cloud computing, should not take three minutes working out whether it will or won’t pour out cash.  But so many factors affect that experience that IT directors have previously been powerless to control the experience.  The era of virtualisation has radically transformed that equation as the IT professional can now isolate, prioritise and manage applications to deliver a fantastic experience to users. 

Finally is the age-old issue of cost.  Every new era of IT has promised much, but at extra cost.  PC networks, client-server computing, server-based computing – all of them demanded an extravagant outflow for the promises of a return tomorrow.   Cloud computing is the very, very first that actually costs less.  By harnessing the scalability and resilience provided through virtualisation, and by using the global networks that now exist, it delivers the massively improved user experience at a lower cost. 

And if you need proof, look at any of the combatants in providing Cloud Computing – Amazon, Microsoft, Google, Oracle – and ask them if they use virtualisation at the core of their infrastructure.  They all do. If the drip, drip, drip of effect of cloud computing works for some of the most popular IT services of today, you can be sure it will seep into mainstream IT soon.

As many of you already know, on August 12, 2008, we experienced a code issue with the updated versions of ESX 3.5 and ESXi 3.5 (Update 2).  The issue triggered a premature expiration of the product license, causing a range of possible problems including the inability to power-on virtual machines on affected hosts.

At that time, I wrote that we would do everything in our power to ensure this type of incident is never repeated.  Since then, we have initiated a major examination and evaluation of our product release and quality assurance processes.  This effort has yielded a number of areas for improvement, and we have already begun making needed changes. 

• We have removed any instances of the type of “time-out” code from all major, minor and maintenance code releases for VMware enterprise products and will not incorporate this type of code in our licensed products going forward.

• We are instituting a range of enhancements to our quality assurance process to place even greater emphasis on standardized, automated quality testing.

• We are driving greater transparency, accountability, and oversight on the entire product release lifecycle with new formal documentation and auditing procedures.

While these are necessary and important steps, they are by no means the sum total of our efforts in this area.  VMware’s products and internal processes have been designed to support mission critical environments, and we take the responsibility that comes with our role as a core element in IT infrastructures very seriously.  To that end, we are taking further actions to improve the overall VMware customer experience going forward.

• First, we will continue to ensure that VMware products are designed to support the reliability, availability and serviceability needs of enterprise production environments.  We are investing significant resources towards improving how our products are installed and upgraded (such as with VMware Update Manager); developing more technology enablers that work to eliminate downtime (e.g. Storage VMotion) and redesigning how licensing and activation work.  To get a sense of where we are pushing the boundaries here, I invite you to register to be part of our beta programs for VMware Infrastructure and other VMware products.

• Second, we will improve and better align our efforts around supporting and communicating to customers in response to product issues.  We have established a company-wide effort intended to streamline and standardize how we reach out to any of our customers affected by a product issue.

• Third, we will continue to provide our customers with the guidance and proven best practices to deploy, scale, and manage VMware across their datacenters.  By enabling our customers to better “operationalize” their VMware environments, we will be helping them to anticipate, plan for, and easily respond to any issues that come their way.  This kind of assistance is already available through our Professional Services and VMware Authorized Consultant offerings.  We are making this kind of information more accessible to all VMware customers with our recently launched VI OPS portal, along with a new “VMware on VMware” program designed to provide insight into how we use our products in our own production IT environment.

We at VMware have always set high standards for ourselves in terms of product quality and customer experience.  Going forward, we’ll continue to raise the bar here.

Thank you for your continued support.

Posted by Carl Eschenbach
Executive Vice President of Worldwide Field Operations

The August 19 announcements from Microsoft are good news for VMware customers.  VMware is now part of the Microsoft Server Virtualization Validation Program Program (SVVP), and is happy to work with Microsoft to deliver better support to our many mutual customers.  Under SVVP, customers running Windows and Microsoft applications on the VMware ESX hypervisor will receive the support they need from Microsoft, in addition to the support they have enjoyed from VMware.  VMware customers can continue to leverage the full flexibility and benefits of using Virtual Infrastructure.  VMware is working with Microsoft to certify ESX in the near future.

VMware Infrastructure provides customers with high levels of continuous service for applications, enabling customers to achieve significantly higher hardware utilization, dynamically balance workloads across physical resources, and move applications in real time. VMware collaborates with Microsoft and other ISVs to ensure that our customers have access to product support consistent with the vendor’s policies for traditional physical environments.

Last night, we became aware of a code issue with the recently released update to ESX 3.5 and ESXi 3.5 (Update 2). 

When the time clock in a server running ESX 3.5 or ESXi 3.5 Update 2 hits 12:00AM on August 12th, 2008, the released code causes the product license to expire.  The problem has also occurred with a recent patch to ESX 3.5 or ESXi 3.5 Update 2.  When an ESX or ESXi 3.5 server thinks its license has expired, the following can happen:

• Virtual machines that are powered off cannot be turned on;
• Virtual machines that have been suspended fail to leave suspend mode; and,
• Virtual machines cannot be migrated using VMotion.

The issue was caused by a piece of code that was mistakenly left enabled for the final release of Update 2.  This piece of code was left over from the pre-release versions of Update 2 and was designed to ensure that customers are running on the supported generally available version of Update 2. 

In remedying the situation, we’ve already released an express patch for those customers that have installed/upgraded to ESX or ESXi 3.5 Update 2.  Within the next 24 hours, we also expect to issue a full replacement for Update 2, which should be used by customers who want to perform fresh installs of ESX or ESXi. 

I am sure you’re wondering how this could happen.  We failed in two areas:

• Not disabling the code in the final release of Update 2; and
• Not catching it in our quality assurance process. 

We are doing everything in our power to make sure this doesn’t happen again.  VMware prides itself on the quality and reliability of our products, and this incident has prompted a thorough self-examination of how we create and deliver products to our customers.  We have kicked off a comprehensive, in-depth review of our QA and release processes, and will quickly make the needed changes.   

I want to apologize for the disruption and difficulty this issue may have caused to our customers and our partners.  Your confidence in VMware is extremely important to us, and we are committed to restoring that confidence fully and quickly.

Thank You,

Paul Maritz
President and CEO
VMware

Posted by Réza Malekzadeh
Sr. Director, Product Marketing & Alliances

It is now just over a month since the very first VMworld Europe and now that the dust has settled it's a good time for us to share some of our thoughts on the show. There were a few key announcements at the event around new technologies we've been working on, all of which demonstrate just how far virtualization has come in terms of being accepted as a mainstream approach to computing.

OEM Announcements

It was especially exciting to see Dell, Fujitsu-Siemens, HP and IBM all standing up to announce that they will be shipping our embedded hypervisor - ESXi 3.5 - as an option with their servers. One of our primary objectives when we announced ESXi was to extend our vision of making virtualization ubiquitous and open up the technology to as many organizations as possible. Consider this - Dell didn't just announce they would make ESXi available with a few of its PowerEdge servers; they announced that they will be making this an option across it’s entire PowerEdge server line. So, once launched, anyone who buys a VMware virtualization-certified Dell PowerEdge server will be able to deploy virtualization easily.

The support of these OEM partners goes beyond representing a huge opportunity for VMware; it also validates the vision for x86 virtualization we had many years ago and the work we have done over the years in effectively creating this market.

Security

There was further endorsement from numerous security vendors around our announcement of VMsafe which was a very significant move for us. For some time now people have been questioning how secure virtual machines are in comparison to their physical counterparts. The argument went (according to some) that if you can successfully attack a hypervisor with a piece of malware then the risks are greater, since that hypervisor has numerous systems running on it.

What we knew was that virtualization can be a fundamentally more secure way of running applications and that it opens up some new approaches to security which simply are not possible in the physical realm. In developing VMsafe we wanted to create an easy way for third party technology vendors - trusted by thousands of customers across the globe - to hook in to our technology and provide extra value for VMware customers.

VMsafe allows third-party security products to gain the same visibility as the hypervisor into the operation of a virtual machine. This allows VMsafe enabled security solutions to identify and eliminate malware that is undetectable on physical machines. This extra layer of protection is achieved through visibility into the virtual hardware resources of memory, CPU, disk and I/O systems within the VM which means every system execution can be monitored. Some journalists have asked whether by 'opening up' our product we are making it inherently easier to attack and more susceptible to malware.

To be very clear, VMware is not opening up its core hypervisor to third-party developers and therefore, potentially, to attack. Instead, what VMware has announced is a secure API; all code running from third parties will be running within a virtual machine, which by its very nature is isolated or 'sandboxed'. VMsafe works on a trust model – so customers will have to select which virtual machines they want VMsafe-enabled security applications to gain visibility into. The key here is that VMsafe-enabled virtual machines will have visibility into VMs and not control over them.

Just as VMware has helped trigger a new wave of innovation in management and storage, we also feel that VMsafe will set off a new round of innovation from security vendors who can harness the power of virtualization to build an entirely new breed of security products.

Automation

There have also been significant developments on our automation and management product lines. At VMworld Europe we gave a preview of what some of our automation and lifecycle management tools, such as Lifecycle Manager will deliver. A month on and we have just announced the general availability of VMware Lifecycle Manager – a product which we feel will help organizations to implement a consistent and automated process for managing the entire virtual machine lifecycle.

We've always believed in innovation, but this has very much been driven by what our customers need and has been delivered in line with the pace at which their deployments and skill sets have grown. VMware Lifecycle Manager is a tool which our larger customers will find makes a real difference to how they can manage their environments. The front end of Lifecycle Manager is completely customizable so customers will be able to build a web-based portal which allows anyone with access to commission, check-out and provision virtual machines. All the policies, permissions and processes associated with virtual machines can be automated and set to run in the background. What you have is a very straightforward, controlled and repeatable way of managing your virtual machine lifecycle.

One of the options we are most excited about in this product is the option to associate chargeback metrics with each virtual machine. This will mean that different business units and / or departments can be charged for the IT resources they use. In the physical realm a concrete cost could be associated with each physical machine - you bought a server, it was yours to run. However, virtual machines have to be treated differently and we have developed this tool to allow costs to be applied to virtual resources. In actual fact, chargeback in a virtual environment is far more granular than it is in physical realms. The cost of buying and building a server doesn't really reflect how much a department uses resources such as storage, networking bandwidth, power etc. With virtual infrastructure, its far easier to get this more detailed breakdown of what resources a machine has used.

Virtual Desktops

We've been making great strides in enhancing our Virtual Desktop Infrastructure offering by developing our existing technology and also through the acquisition of companies like Propero. At VMworld we previewed a couple of technologies that are incredibly exciting.

The first of these technologies is Offline Virtual Desktop Infrastructure. Users can check out personalized desktops to a notebook, use them in offline mode and then check back in to the desktop running in online mode. This really takes VDI beyond the datacenter and means users can have a single image to use whatever machine they are using. It also means VDI can be as flexible as our customers require.

One of the issues with VDI that our customers had brought to our attention was that a large VDI deployment would require a lot of storage to support it; if you are supporting five hundred instances of Windows XP that is a large amount of bulky operating systems to have sat on your shared storage. This is why we have developed our scalable virtual image technology in a tech preview. This allows desktop images to be published to hundreds or even thousands of virtual machines without the need for each image to be individually stored; instead we can clone the virtual machine image. This can reduce storage requirements by up to 90% which we believe will really help the uptake of VDI in general. We take the desktop virtualization market very seriously - when you think about how many PCs are out there we think customers will always be looking at better ways of managing enterprise desktops, and in VDI we think we have a great solution.

So between these four areas where we have made announcements - OEMs, Security, Automation, Virtual Desktops - we think we have really demonstrated our development efforts. Our OEM deals demonstrate how we are making virtualization more accessible and easier to deploy; VMsafe shows that we take security incredibly seriously; our Automation products reflect our desire to make virtual infrastructure easier to manage; the VDI tech previews show that VMware is determined to make virtual desktops easier to deploy, manage and use.