The Console

VMware's Executive Blog

Wed, 25 Oct 2006

Posted by Dan Chu
Senior Director Developer and ISV Products and Technology Alliances

In April of this year, VMware announced that we were making our virtual machine disk format, VMDK, openly available and freely usable to anyone who wanted to do so. Since then, over 2000 vendors and developers have requested to review and use our VMDK specification. The virtual machine disk format is a critical specification for customers and the industry because value-added solutions that manipulate virtual machines such as patching, provisioning, backup, and security are dependent on it.

Last week Microsoft announced that it too is moving to make its virtual machine disk format, VHD, more open. Previously VHD had been covered by a much more restrictive license. We are glad that Microsoft is making VHD more freely usable by third parties. The ecosystem has invested broadly in VMDK, but it is good that VHD now has the same accessibility.

One highly related area we are concerned about is that we’ve seen Microsoft beginning to put restrictive terms on the use of published VHDs. Specifically, it seems that Microsoft is starting to restrict use of their VHDs to MS Virtual Server and Virtual PC only. In contrast, there are over 300 VMDK-based virtual appliances available on VMware Technology Network (ranging from Oracle databases to CRM packages to firewalls to email security solutions) that are freely usable by all regardless of platform or product.

If Microsoft constrains software licensing of the content within VHDs so that the VHDs can only be run on Microsoft products, then there won’t be any real openness or interoperability for VHDs. We hope that Microsoft is committed to interoperability and open implementations for VHDs, and that the chokepoint isn’t simply moving from one prohibitive licensing constraint (VHD format licensing) to another (VHD software licensing). Microsoft should appreciate (and customers and partners have been very clear about this) that a closed system based on licensing restrictions which lock customers into Microsoft’s products and formats isn’t acceptable.

posted at: 11:59 | reply to the console | permanent link

Fri, 22 Sep 2006

Posted by Bogomil Balkansky
Director of Product Marketing

The datacenter has typically been a quiet, and face it, boring, place: rows upon rows of identical computers and the droning sound of air conditioning – not exactly the place where journalists look for spectacular drama. And you want to keep it that way – if things are working as they should, no one needs to know what is going on in the datacenter, or that such a place even exists.

So why has the datacenter come out of its anonymity, and is suddenly filling the pages of the New York Times and the Wall Street Journal? And why are huge billboards on highways and airports touting the concern about computer energy consumption and cost? The truth is that rising energy prices have hit the datacenter as hard as they have hit Detroit. Modern servers have increasingly become the computing equivalent of gas-guzzling SUVs. While a typical server 10 years ago consumed 100W of power, the average server today consumes four times as much. Servers would use about 30% of their peak electricity consumption while sitting idle, which is often more than 80% of the time. Imagine your SUV going through gallon after gallon of gas while sitting in the garage. And to make things even worse, the density of servers per square foot has doubled at the same time – from 7 servers per rack to 14 servers per rack. The overall power density of the datacenter is increasing by 15% per year. All the electricity consumed by servers is transformed into heat – so to prevent data centers from turning into hot houses, about 125% more electricity is consumed by the cooling equipment.

IDC calculates that the total power and cooling bill for servers in the US stands at a whopping $14 billion a year, and if the current trends persist, the bill is going to rise to $50 billion by the end of the decade. The growth of datacenter energy spending far outpaces the rate at which IT budgets grow, dangerously crowding out other vital IT initiatives and projects.

Not that virtualization is a panacea for every IT woe, but it can definitely help overturn that dire forecast. One of the mainstay use cases of virtualization – server consolidation and containment – allows customers to “squeeze” multiple workloads on the same server. There is a flow through effect from needing fewer physical servers – it means that VMware customers need less space in the datacenter, and less electricity and cooling. We estimate conservatively that for every workload moved from a physical to virtual environment, customers can save about $290 in electricity costs, and about $360 a year in cooling costs. The more important thing is that these savings accrue year after year. For example, VMware customer Provident Bank reports cutting power consumption by 13,000 watts.

Beside the company bottom line effect, there is something to be said about the environmental impact of virtualization. The $650 per virtualized workload represents 8,000 kWh of electricity saved. With more than 1 million workloads running in VMware virtual machines, the aggregate power savings are about 8 billion kWh, which is more than the heating, ventilation, and cooling electricity consumed in New England in a year. With results like that, your datacenter won’t mind some attention from the press.

posted at: 15:36 | reply to the console | permanent link

Wed, 13 Sep 2006

Posted by the Engineering Performance Group

As virtualization becomes commonplace in the industry there is increasing interest in measuring the performance of virtualized platforms. Plenty of benchmarks exist to measure the performance of physical systems, but they fail to capture essential aspects of virtual infrastructure performance. We need a common workload and methodology for virtualized systems so that benchmark results can be compared across different platforms.

There are a number of unique challenges in creating sound and meaningful benchmarks for virtualized systems:

• Capture the key performance characteristics of virtual systems
• Ensure that the benchmark is representative of end user environments
• Make the benchmark specification platform neutral
• Define a single, easy to understand metric
• Provide a methodical way to measure scalability so that the same benchmark can be used for small servers as well as larger servers

Let's take a look at these in turn.

Capture the key performance characteristics of virtual systems. Users compare platforms based on their specific needs -- for example a user running a web server will compare the number of web requests that can be served by each platform, while a DBA will be more interested in the number of database transactions or simultaneous database connections. In the non-virtualized world users typically bind a single application to a single machine, and benchmarks have been developed to provide metrics for important application categories. For my examples of web servers and databases users could look at results for the industry standard SPECweb2005 and TPC-C benchmarks, respectively. Such benchmarks run the application to a point where some resource (usually CPU) is saturated and record the performance metric while respecting quality of service measures such as response time. Users not only compare platforms using the metrics provided by these benchmarks, but over time they build up expertise allowing them to relate their particular environment to published benchmark scores.

However, in a virtual environment, the typical usage of a machine is different from what is common on physical machines. One of the key benefits of virtualization is the ability to run multiple virtual machines on the same physical machine to increase the utilization of server resources. Multiple virtual machines running different operating systems and different applications with diverse resource requirements can all be running on the same machine. Moreover, these applications are typically not bottlenecked on any one resource, and have different (often conflicting) response time requirements. Running single application benchmarks one at a time and then aggregating their metrics might appear to be an easy solution, but that approach doesn’t work. Overall performance can be negatively impacted by competing resource demands among the workloads or positively impacted by optimizations such as transparent page sharing. A good virtualization benchmark must include multiple virtual machines running simultaneously.

Ensure that the benchmark is representative of end user environments. Which workloads should run within the virtual machines in the benchmark? This is a difficult question as users run a wide range of guest operating systems (e.g. Windows, Linux or Solaris), virtual hardware configurations (32-bit, 64-bit, 1, 2 or 4 virtual CPUs) and applications in virtual machines. Any workload included in the benchmark must be representative of end user applications, especially in terms of their resource usage. For many virtual environments, CPU utilization of the workloads is an important factor in the overall performance of the system, but so are memory, storage and network I/O. Any benchmark that aims to measure the performance of virtual environments is incomplete if it does not address these resources. A virtualization benchmark must take into account existing customer use cases and future trends in hardware and software.

Make the benchmark specification platform neutral. Care must be taken to ensure that the benchmark specification does not depend on any platform specifics. We’d like to be able to use the benchmark to answer common customer questions such as “What’s the benefit of dual-core (or quad-core) over single-core processors?”, “How does my storage hardware affect the performance of my overall system?” or “What’s the performance difference between hosted virtualization products (like VMware Server) and bare-metal virtualization products (like VMware ESX Server)?”. Another aim of this benchmark is to drive improvement in future platforms and we would not be able to accomplish this if the benchmark was tied to any specific platform.

Define a single, easy to understand metric. Any good benchmark will have a single, simple metric so that it’s easy to compare different platforms. Secondary metrics can be used to give additional information, but users will base their comparisons on the primary metric. For a virtualization benchmark, should latency or throughput be the primary metric? Or should the load on all the virtual machines be held constant, and CPU utilization used as the metric so that the systems that are able to handle the load with the lowest CPU usage are deemed best? Should the CPU utilization be capped at some limit or should the workload be allowed to saturate the server? How are quality of service constraints factored in? Considerable experimentation is required to determine the best choices for the benchmark design and hence the validity of the benchmark metrics.

Once the metric for each component workload is defined, the next question is how to aggregate them. The aggregation must be done carefully as the units of the underlying workloads can vary widely and we don’t want a single workload unfairly influencing the final metric. The aggregation should also be meaningful with regard to making the benchmark representative of what end users really run. In addition, the metric for a new benchmark must be easy to reason about, make sense to end users, be easy to compute and reflect underlying platform differences.

Provide a methodical way to measure scalability. One of the key benefits of virtualization is being able to consolidate workloads in a scalable manner onto machines. It’s important that the benchmark be able to run on a small two CPU system as well as on the large multicore, multisocket system of tomorrow and provide a meaningful measure of the relative work that can be performed on the two systems. Besides CPU, platform differences in storage and networking hardware can also affect scalability of the system and need to be captured by the benchmark.

In this article we’ve discussed some of the main design challenges for a virtualization benchmark. Turning the design into an easy-to-use benchmark kit brings up additional practical considerations. These include issues such as timing in virtual machines, orchestrating the startup of multiple virtual machines running simultaneously, and determining the right measurement window in the face of bursty workloads. Any approach to creating a benchmark for virtualization must address all these challenges. Creating a benchmark is easy, but creating a credible benchmark that provides a meaningful metric, that measures both workload overhead and scalability, that is representative of end user environments, that cannot be easily defeated, and that is broadly applicable -- is a hard problem!

The benefits of solving this hard problem are great. Having an industry standard way of comparing virtualized solutions will allow users to make more informed decisions regarding the entire stack of virtualization technology. Such a standard can also drive improvements in future hardware and software, again benefiting the industry. For these reasons, VMware is committed to solving this problem. For a while now we've been working on just such a benchmark. We’ve been talking to many of our customers and partners and doing lots of experiments to develop a sound design and methodology. We're referring to this benchmark as VMmark (for Virtual Machine benchMark) and we plan to present it at VMworld 2006 this November (session TAC9668). This is part of VMware’s larger effort to promote open standards and formats within the industry. We don't intend for this benchmark to become a VMware product. Rather, we’d like to work with others to make this benchmark an industry standard and thus provide users with the standardized method of comparing platforms they have come to expect from enterprise software.

posted at: 00:00 | reply to the console | permanent link

Mon, 14 Aug 2006

Posted by Srinivas Krishnamurti
Director of Developer Products and Market Development

In June 2005, we launched our VMware Technology Network where in addition to a subscription offering for developers and a new online resource dedicated to virtualization – www.vmtn.net, we introduced the concept of virtual appliances. Virtual appliances are pre-built, pre-configured and ready-to-run software applications packaged with the operating system within virtual machines. We believe that application vendors should choose the best operating system for their application and then deliver the software service to the end user, ready-to-run. This puts to end to the significant amount of time IT administrators spend installing and configuring software, which is really nothing but a cover charge for using software. We started with about 6 appliances from vendors such as Oracle, BEA, etc., but soon we received positive feedback from both customers and partners alike.

With the release of the royalty-free VMware Player, we saw a huge up tick in the number of virtual appliances developed. The community at large was creating appliances that they thought were useful and better yet, they were nice enough to share them with others via VMTN. We noticed the passion and thought that it would be a cool idea to host a competition to foster innovation and reward folks for their work. That’s how UVAC was born – we announced it on Feb 27, 2006. We reached out to industry leaders from different walks of life to be judges and the final list was fantastic.

Hundreds of contestants registered. Folks on Slashdot thought the contest was a cool idea. The community forums were buzzing. When all was said and done, over 170 appliances were submitted. We immediately uploaded them to VMTN so the community could vote for the Community Choice. We then uploaded all the submissions onto USB drives and sent them to the judges so they could pick the winners.

Over 75,000 downloads later, the community picked its winner:

• Community: Hermes Memcached Virtual Appliance Prabhakar Chaganti of Alpharetta, Ga., is a high-performance, distributed memory, object-caching system that provides a simple and fast caching mechanism to speed up database-driven applications.

Many email exchanges later, the judges decided on their winners:

• Collegiate: Open Network Registrar by Zachary Colgan of Potsdam, N.Y., is a drop-in administration tool that simplifies the simultaneous administration of BIND and DHCP.

• Consumer: FreeNAS by Olivier Cochard of Paris, France, is a Web-based NAS server operating system based on FreeBSD 6.1 that provides network-attached storage services such as CIFS, FTP, NFS, SSH, SCP, RSYNC and AFP so users can access storage with a diverse set of protocols.

• Developer: eBrainPool by Jatin Golani of Mumbai, India, automatically shares development environments through the P2P eBrainPool network so developers can search for and connect to other shared development environments that are available to the P2P sharing application. With this P2P network of environments it is no longer necessary to have every developer tool and every OS combination because developers can now share environments with others for specific tasks.

• Server: Enterprise Encryption Server (EES) by Carl Bourne of Henley on Thames, United Kingdom, is an automated OpenPGP-based data encryption and decryption utility that makes it easy to deploy and manage centralized OpenPGP-compatible encryption resources. Send an unencrypted file to the EES virtual appliance (via FTP or through a shared directory) and it will send back an encrypted version ready to be transmitted securely over the Internet.

• Third prize: Sieve Firewall by Michael C. Jett of Senath, Mo., and Kennieth A. Goodwin of Paragould, Ark. Sieve Firewall makes it easy for Windows users to set up and use a transparent bridging firewall without having to learn Linux. The virtual appliance is managed by a Windows GUI application. The resulting XML configuration files are loaded to the Sieve Firewall virtual appliance and filtering can be up and running in minutes. Because the firewall is completely invisible to the outside world, it is not vulnerable to attacks that target more traditional firewalls. Not only can this appliance be used to create safe home networks, it can be used to manage and prioritize traffic in a multi-department, complex business network.

• Second prize: Trellis NAS Bridge by Andrew Macdonell, Michael Closson, Paul Nalos and Paul Lu of Edmonton, Alberta. The Trellis NAS Bridge Appliance makes it simple to access files across the network regardless of location, operating system or type of file sharing technology (SSH, NFS, SMB, etc.). It simplifies file access while maintaining security.

Drum roll please…

• First prize: HowNetWorks by Mikko Hiltunen, Erno Kuusela, Joachim Viide, Mika Seppänen and Jani Kenttälä of Oulu, Finland. HowNetWorks is an always-on, all-in-one, personal network troubleshooting console for those tired of the “laborious work” of network debugging. No more writing complicated sniffer filters, no more searching for ways to reproduce failures. HowNetWorks takes the next step in sniffer evolution—simply fetch the relevant data and throw it to your favorite analyzer

The judges put in a lot of work to narrow down the list of appliances. Many of the appliances were discussed at great length by the judges and in the forums by the community. You could tell that contestants put in a lot of time and effort in building these appliances but we could only pick eight winners. We decided to hand out Honorable Mentions to other appliances that didn’t win but ones that were picked as finalists. Thanks to the contestants, judges and the community at large for making this an exciting challenge. Drop me a note if you have ideas on how we can do this better next time.

Next time I will write more about all the rage about virtual appliances and why VMware’s virtual appliances are the real deal.

posted at: 00:00 | reply to the console | permanent link

Mon, 07 Aug 2006

Posted by Srinivas Krishnamurti
Director of Developer Products and Market Development

Ever since Apple announced that they are going to switch to Intel x86 chips, phones have been ringing off the hook. Everywhere we went - customer visits, tradeshows, seminars, VMworld, neighborhood block parties - everyone was asking, "Hey, what are your plans for supporting Macs?" Some folks went a step further to petition us to support OS X.

Traditionally we don't talk about products that are still under development for a variety of reasons. But word got out that we have something "working in the labs." And we started getting more calls along the lines of, "Hey, we know you have it so when are you going to announce it?"

So we bucked the trend! Today we announced our plans to build a new product for Macs. We have a bit more work to do -- bug fixes, installer cleanup, etc. -- before releasing the beta. We are hopeful to have something for you to try out in the next few weeks. If you are interested in getting early access to the beta, please register here.

This product will allow you to create and run virtual machines on OS X, which means that you can run Windows, Linux, Solaris, FreeBSD, Netware and a slew of other operating systems on OS X without rebooting. You can switch between OS X and your favorite operating system similar to how you can switch between two applications. You can even share files between the two operating systems with powerful drag and drop features. We firmly believe in freedom of choice when it comes to what OS you want to run. With this new initiative, we extended our already vast array of supported platforms.

So if you are like my Danish colleague, Rene, who uses his own Apple laptop at work for most applications but has to switch to a work PC to access some applications, you can now safely shelf your PC and make the Mac your primary and only desktop. Just create a virtual machine, install Windows (or whatever you run on your PC) in it and run it alongside OS X. Now you're talking!

We also agreed early on in the design process to make sure that your experience of running a virtual machine is very close to native, in terms of performance and use of hardware/devices. To that end you will notice excellent performance when running a virtual machine on OS X. Since most Macs now have Intel's Duo with two cores, we added Virtual SMP capabilities so that you can assign more than a single CPU to any virtual machine to gain additional performance. One of the cool things that I like about my Apple (albeit an older one with PowerPC chip) is the simplicity of doing any multi-media work. I can record movies or use iChat quite easily without the need to buy additional software. We wanted to make sure you have access to all those devices from your virtual machine as well so we spent a lot of time on making sure devices work well. You can use USB 2 devices like video cameras, etc.

One final note about this product: virtual machines created with this product are fully compatible with the latest release of other VMware products, including Player, Workstation, Server and VMware Infrastructure and vice versa.

Besides just the product features I described above, now Mac users are part of the VMware family. One of the cool things about being part of the family is that you can access the directory of virtual appliances at the VMware Technology Network. There you will find over 250 virtual appliances, which are pre-built, pre-configured and ready-to-run software applications, all packaged within virtual machines. No more installing and configuring software. Just download a virtual machine and run it with a VMware product. Getting a blogging server set up has never been easier!

We are very excited about bringing this product to market and look forward to your feedback during the beta process. I don't know about you but I'm going to the store now to get me a MacBook because I love my Mac and want to have a PC, too.

posted at: 00:00 | reply to the console | permanent link

Tue, 18 Jul 2006

Posted by Brian Byun
Vice President of Products and Alliances

Virtualization is fast becoming a ubiquitous layer of software that transforms how applications are developed, deployed and managed in IT environments of all sizes. Not long ago, Diane Greene explained the need for three major areas of industry standards around virtualization that are required to support customer choice as virtualization adoption becomes pervasive. These standards will enable customers to choose their virtualization software, management tools and applications, and associated OS on merit-based criteria: quality, functionality, and price rather than traditional licensing and arbitrary vendor-specific lock-in rules.

Yesterday, Microsoft announced that they would work with XenSource to allow para-virtualized versions of Linux to run on future Windows hypervisors. Para-virtualization is an emerging technology that requires a modified guest operating system to call an API on the underlying hypervisor, in order to optimize the performance of the guest OS when it is run on a virtualization software layer. In light of this news, it's worth re-visiting some current developments in standardizing the interface between the operating system and the hypervisor.

First, let's take note of a few ironies of this recent Microsoft/XenSource arrangement:

• It's a one-way street that favors Microsoft and Windows running Linux. The arrangement will allow Linux to run on future Microsoft hypervisors through translated calls to the hypervisor when Windows is controlling the hardware, but not the other way around; i.e. there is no mention of Longhorn optimizations or "enlightenments" being ported to Xen or licensed to XenSource to enable a Xen hypervisor to run full optimizations with Longhorn OS.
• XenSource, in diverging from its open source and Linux virtualization roots, is enabling the commercial interests of Windows and building to proprietary Windows API layers. It stands to reason that in order to protect Windows from GPL contamination, XenSource will need to undertake a lot of non-GPL development to translate and buffer the Linux kernel from Windows hypervisor interfaces; and nothing that Microsoft licenses to, or develops with, XenSource is GPL and can be used directly by the Xen or Linux communities and commercial distributions.
• The timing is rather notable. It's odd to trumpet future interoperability for the Windows hypervisor whose first release is roughly two years away or more, while the Linux hypervisor interfaces are still being actively discussed in the open source community.
• In the end, it remains to be seen how relevant this one-way "bridge" to Longhorn will be with Linux customers and community. After all, no one adopted Linux in order to run it on proprietary Windows interfaces, which is what's contemplated here. In addition, there is a notable absence of support for this scheme from major Linux commercial distributions, OEMs and customers.

Impact of the OS-Hypervisor Interface

This announcement gives reason to examine the impact of the OS-hypervisor interface more closely. Ideally, any hypervisor should be able to run optimally with any operating system and there should be no proprietary license required to do so. Under the announced arrangement, XenSource is licensing a Microsoft owned OS-hypervisor interface but not the other way around. Microsoft wants to make Windows run optimally only on Microsoft's hypervisor but are happy to let other operating systems such as Linux run on top of their hypervisor. Clearly Microsoft views their control point as moving to the hypervisor.

VMware hopes there will soon be a standard Linux interface for para-virtualization, which will simplify and standardize how Linux is supported on various hypervisors, including VMware and Xen. VMware is actively working with the Linux kernel community to develop an open interface so that the Linux kernel can run natively and efficiently on a choice of hypervisors. Such an interface would also be available to any operating system. VMware has made its initial proposal for such an interface available to the Linux community and is pursuing Linux and hypervisor interoperability not as a commercial arrangement, but within the open, transparent, and merit-based multi-vendor approach that is the hallmark of the Linux kernel community.

The industry is realizing that the x86 virtualization trend provides a unique point in time where customers can exercise control and freedom of choice of OS, applications, hardware vendors and virtualization stacks, and where no one vendor dominates and controls. This results in pure value-based competition and opens up a world where any OS or virtualization approach can potentially run any other OS. It's up to industry participants and customers with buying clout to fully enable this vision. In this new world, we recommend that customers demand unfettered, open standards and full bi-lateral interoperability from all your vendors before expanding or doing business with them.

In the end, I believe most customers will run their Linux distributions on hypervisors that use open standards and do not have license lock-in to proprietary interfaces. This is the approach VMware takes. Every key interoperability interface VMware implements in its products is available on license-free and open terms to its partners as well as competitors.

Stay tuned to the "virtualization standards" channel as we'll be reporting back often. Do let me know if you have comments for VMware on its approach to industry interoperability based on open standards.

posted at: 00:00 | reply to the console | permanent link

Wed, 21 Jun 2006

Posted by Dan Chu
Senior Director Developer and ISV Products and Technology Alliances

The Virtual Software Lifecycle: VMware for Development to QA to Production

In the two days since I began to write this blog, we've had tens of thousands of downloads for VMware Server and VMware Player – adding to the millions of downloads of these products since their market debut (February 2006 for VMware Server beta and October 2005 for VMware Player). These are unheard-of download rates from an infrastructure software standpoint – virtualization has really become mainstream. Importantly, most of these downloaders have never used VMware software before!

Further, we are finding that the leading usage for VMware Server is development and test. More than 52% of 12,000+ VMware Server users that we surveyed indicated that development and test is their primary use case. Just as the newly released VMware Infrastructure 3 transforms static farms of servers, software, storage, and networks into flexible resource pools, VMware Workstation, Server, and Infrastructure together transform the software lifecycle of development, test, and QA for millions of users.

Every IT and development organization shares a similar software lifecycle – development and test environments are eventually deployed as production environments, which iterate back into dev, test, and QA, the cycle repeating itself with every release. I hear the same issues from every organization: There are a huge number of development and test environments – in many companies, 2-3 machines in application development and test for every server in production is the norm. Moreover, processes are overly manual and too resource-intensive – for a typical development organization, the overhead for setup, provisioning, and teardown for test/QA/development environments accounts for more than 50% of time expended in the entire development and test cycle!

These issues are driving what has become a consistent request from our customers and systems engineers: they'd like to see VMware streamline and optimize the various stages of the software lifecycle, with capabilities such as automating and managing operations for a specific stage (e.g., testing), as well as processes that span stages, such as configuration capture and management.

At a Customer Advisory Council earlier this year, our customers repeatedly described the need for virtual machine libraries, automated and self-service provisioning, and dev/test workflow leveraging virtualization. For example, a senior IT architect from one of the largest media companies in the world expressed these needs at a customer feedback session, repeated himself at dinner that night, and then told me again when we saw each other two weeks later. Our customers know what they are talking about, and they're passionate about it!

The Virtual Software Lifecycle

Akimbi

At the same time, we've heard from customers who use VMware solutions across their software lifecycle that the new capabilities from Akimbi Systems provide a perfect complement. The director of IT at a global retailer told me that the Akimbi capabilities for virtual lab automation were a great bridge from their use of Workstation for developer desktops to their broad use of VMware Infrastructure (including ESX Server and VirtualCenter) for their production applications.

While working on our software lifecycle capabilities internally, we've also been working with Akimbi for more than a year (Akimbi was a Technology Alliance Partner and participated in and leveraged VMware's Community Source program). Most importantly, their technical leaders – in particular Wilson Huang, Akimbi co-founder and VP of engineering – spent a lot of time with our senior engineers, gaining a sizable amount of mutual respect in the process. We also think very highly of James Phillips, CEO of Akimbi, whose vision and understanding of the software lifecycle matched up very closely with ours.

With this shared respect and vision on both the technical and business sides, combined with the clarity of market demand, we agreed to join forces. Last week we announced that VMware acquired Akimbi. The Akimbi team will join with ours – and move south a couple of Highway 101 exits to Palo Alto. We are especially excited about the key leadership roles that the Akimbi founders will play: James will spearhead VMware's overall dev/test and software lifecycle business, and Wilson will lead and drive our R&D efforts focused on the software lifecycle. It promises to be an exciting and eventful summer - and beyond!

Let us know if you have thoughts about or interest in this solution.

posted at: 00:00 | reply to the console | permanent link

Tue, 02 May 2006

Posted by Jeff Jennings
Vice President of Systems Engineering

When I joined VMware six years ago, the only product we sold was Workstation. The focus market was already well defined: developers, testing, and Linux enthusiasts. But necessity being the mother of invention, a few customers came up with ideas of their own. One of the more interesting use cases came up shortly after the Workstation 2.0 release. We received a number of inquiries from law enforcement agencies, including the FBI. "What is the best way," they asked, "to convert a physical drive to a virtual machine?" The reason, they explained, was for computer forensics - studying hard drives for incriminating evidence. The officers' existing approach involved making an image, restoring the image on a research computer, and then finally searching for evidence. Because of the limited number of computers, the process was overly complex and slow. What they needed was an automated process for converting a physical drive to a virtual machine. What they needed was the P2V product, which would not come for another two years, but I remember being impressed by these customers' inventiveness and proactiveness in contacting us.

Fast forward to the current day and customers are still showing us new and exciting ways to use our products. I would say the most recent is VDI or Virtual Desktop Infrastructure. You are probably familiar with the big three uses cases of virtualization: server consolidation, development & testing, and business continuity. But there is a new solution on the block, and it is here by customer demand. VDI brings a hosted desktop solution to the VMware portfolio focused on enterprise manageability and security.

The first request I encountered came from a large software development company in California. They had three key goals: support offshore development sites, reduce the number of operating system licenses (they only needed licenses for operating systems loaded in memory for a VDI solution), and support a thin client option in their home offices. Of these, supporting the offshore sites was the critical need. This was back in the ESX Server 1.x days and the customer felt that if they could get 8 desktops per two-way server, they would be happy. We were able to achieve that ratio, which was a challenge because our memory management was not nearly as advanced as it is today, but at the time they choose not to go forward. Shared storage was an integral part of the solution and they were not willing to commit to purchasing another array. The individuals involved, however, recognized that this would fundamentally alter how they managed their desktop environment.

They have proven to be quite prescient. Prudential UK, for example, implemented a VDI customer service center. The desktop servers are located in the UK and the customer service representatives are in India. This allowed them to bring up the customer center in only 4 months. Interestingly, the set of target applications would not work in a traditional terminal services environment. VMware ESX Server allowed them to run out of the box, without rewriting. Prudential UK cites the ability to provide "applications anywhere" as a major benefit of VDI.

This is not lost on our other customers in Europe. Recently I have seen numerous enquiries regarding business continuity. Specifically these customers are looking for a way to provide a complete corporate environment to their employees' homes in the event of a bird-flu pandemic. VDI would give them the ability to continue in business if public health concerns prevented their employees from coming to work. That is certainly a highlight of VDI - the ability to provide a complete enterprise environment anywhere you have a robust network, yet still have centralized management.

And there you have it: from fighting crime to managing a potential pandemic, VMware virtual machines have you covered. What will our customers think of next?

posted at: 00:00 | reply to the console | permanent link

Fri, 21 Apr 2006

Posted by Jerry Chen
Director of Enterprise Desktop Platforms and Solutions

Today VMware announced the Virtual Desktop Infrastructure Alliance, which may mark an inflection point in desktop virtualization. First, let me give some background and history on Virtual Desktop Infrastructure, or what we usually abbreviate as VDI (you can imagine that we use a lot of three letter acronyms that start with the letter "V"). The interesting thing about VDI is that it is largely a customer driven solution. As early as 2002 I've seen customers use VMware virtual infrastructure software to host desktops in their data centers. Basically, an administrator creates a desktop virtual machine (e.g. Windows XP) and runs it on a server in the data center. The actual desktop end user connects to the virtual machine using a remote display protocol, most commonly RDP. Users can connect to their desktops from any device that supports RDP, including thin clients and even regular PCs. All you need is a network connection and you can access your desktops. VDI essentially turns your entire desktop into a hosted application like Salesforce.com. More and more applications are moving from client hosted to server hosted, and VDI extends that paradigm by transforming a PC into a hosted application.

The most frequent question I get about VDI is "How is this different from terminal services or thin client solutions of the past?" The major difference between VDI and these other offerings is the fact that users now are accessing their own individual PC environment instead of accessing a shared application environment. With Virtual Desktop Infrastructure, users interact with a real desktop that looks and behaves like their normal PC. An added benefit of using virtual machines is that each and every user is isolated from each other resulting in improved security.

There are probably two common ways to set up Virtual Desktop Infrastructure:

Static Desktops. In its simplest incarnation, Microsoft Windows XP desktop virtual machines run on VMware virtual infrastructure and are accessed remotely from a PC or thin client using RDP. This is the basic or static one-to-one; every user accesses the same desktop virtual machine by initiating an RDP connection to the same IP address.

Pooled Desktops. In a more sophisticated deployment, users log into a portal, are authenticated and then assigned to a virtual machine from a pool of generic virtual machines. In this setup, a system admin would create a batch of desktops or groups of different desktops and then assign users to a group based on pre-determined business rules. For example, a developer would log into any one of a thousand developer virtual machines while a marketing manager would log into a different virtual machine build with different applications.

These more complicated offerings are a major area of emphasis for the VDI Alliance. VMware is working with the different members of the Alliance to create tested and integrated offerings for customers to deploy for their desktop hosting projects.

Virtual Desktop Infrastructure gives customers increased control and flexibility of their PC environments. PC management and maintenance becomes easier because it eliminates the need to walk to a user's cube to troubleshoot a problem and it enables PC upgrades with a click of the mouse by increasing the CPU and memory resources to a virtual machine. I've spoken to a couple of customers who are planning a major desktop and OS refresh and are planning on repurposing their old PCs as dumb clients and moving the desktop to a virtual machine on the server. By turning their old computers into effectively VDI terminals, they can extend the life of their IT assets while still upgrading their users to a new virtual computer running on Virtual Desktop Infrastructure.

How does Virtual Desktop Infrastructure compare to our other desktop products like VMware ACE, VMware Player, and VMware Workstation? These three products run virtual machines locally on a user's laptop or desktop. In comparison, VDI runs a desktop virtual machine on the server. In particular, ACE is the perfect complement to VDI. VMware ACE creates a secure desktop virtual machine that runs on a laptop or desktop while VDI creates a desktop virtual machine that runs securely on the server. One VDI Alliance partner called VDI "ACE-online." Enterprises can now use virtual infrastructure to manage online and offline users.

Use Cases

Why does Virtual Desktop Infrastructure make sense? The two major benefits of VDI are centralized management and security. Desktops today are the most underutilized and difficult to manage IT asset in the enterprise. The average PC sits idle for most of the day and there are gigabytes of sensitive data sitting at the edge of the network, unprotected on PC hard drives. By centralizing all of these computers into the data center, an IT administrator can increase utilization and improve security. Add in the power and flexibility of virtualization such as VMotion technology to manage resource demands, and you have a very compelling management solution.

The major VDI use cases that I see from customers:

• Off shore Development. More and more QA, test, and some basic development is being outsourced or off shored to third parties or in-house foreign development offices. Enterprises need to provide developers independent desktops but want to keep IP secured in the US.
• Outsourced Call Centers. A close cousin to off shore development is outsourced call centers. Some legacy applications do not work well in a shared services environment and certain regulated industries require strong user isolation. Both of these problems can be solved using Virtual Desktop Infrastructure.
• Branch Office Consolidation. Branch office locations like a retail bank offices with a few tellers or a retail store with a couple of point of sale PCs are also strong candidates for Virtual Desktop Infrastructure. There is no local IT staff to call when the PC goes down and there is sensitive customer data on the computers. By taking the PC out of the branch office and moving it into the data center, an IT admin can give better support to the remote locations while improving security.
• Desktop Disaster Recovery. Disasters like Hurricane Katrina have spawned hundreds of DR projects. If something should happen to a company's office, how can the enterprise keep employees productive? If the desktops are now stored in virtual machines on network storage, a company can use existing data center DR solutions to support desktop DR.
• Alternative Workspace. Related to Desktop DR is the alternative workspace use case or the pandemic scenario. Don't laugh, but just like companies are funding DR projects, they are funding SARS and Avian Bird Flu projects. I was talking to a customer in Asia last week that confirmed when SARS hit their city a couple of years ago, all their workers were quarantined at home. With VDI, employees can access their desktops from home.

Whatever the use case for Virtual Desktop Infrastructure, it's amazing to see how enterprises are beginning to use virtualization to solve problems beyond server consolidation. As more of our customers begin to standardize on virtualization they are leveraging VMware virtual infrastructure to solve other pressing IT problems such as PC management. I don't think it will be long before we see companies benefit from the value of virtualization from the desktop to the data center.

posted at: 00:00 | reply to the console | permanent link

Mon, 10 Apr 2006

Posted by Jack Lo
Senior Director of R&D

There was a flurry of press releases around virtualization this week at LinuxWorld. The topic is certainly very hot (although the firefighters who showed up at the show weren't there for virtualization, they were there for the smoke at the Unisys booth.

I was really excited to see a growing recognition that virtualization is a lot more than just server consolidation. Virtualization isn't just about carving up your server to run multiple OSes and save money. There are so many amazing things you can do with virtualization, and it's fun to see others realizing the kind of impact that virtual appliances can have.

I did hear a lot of discussion about how the virtualization hardware from the CPU vendors will greatly improve performance. I guess it's natural for people to assume that HW will be faster than software. But the misconception is that HW support will make everyone's virtualization software faster than VMware's. The first-generation hardware assist is more of a functionality enabler -- it makes it easier for people to write an x86 virtual machine monitor. Also, all vendors, including VMware, can take advantage of the performance features of the new HW. But it's still really hard to make a high-performance vmm like VMware's.

One of the most entertaining sessions at LinuxWorld didn't really have anything to do with virtualization, though. There was a panel on "The Death of the Enterprise Software Business Model: How Startups are Leveraging Open Source to Change the Model". Marc Fleury from JBoss was quite the crowd pleaser with his humor and potshots at Microsoft, IBM, and Sun. All of the panelists held the view that open source software is a superior model to proprietary software. But time will tell... A couple of their key points were that 1) open source make it really easy for customers to try before they buy, and 2) open source software is high-quality because developers are much more careful, knowing that the whole world is watching the code they write. These are both valuable outcomes, and VMware users and Community participants are getting those same benefits through VMware's free Player and Server products, and a strong code review process both in the Community Source program and in our internal engineering organization.

posted at: 00:00 | reply to the console | permanent link

Thu, 30 Mar 2006

Posted by Dan Chu
Senior Director Developer and ISV Products and Technology Alliances

This past week I heard about a software vendor that attributed 40% of their support costs to be essentially operating system support.

In a recent survey by Information Week, the #1, #3, and #5 issues slowing down deployment of Linux applications are:

   #1: complexity of deployment, technical expertise needed
   #3: multiple Linux versions
   #5: hardware support

Over 40% of the x86 server market worldwide (> 2.5 million new servers per year) is IT infrastructure.

What do these have in common? They are all compelling drivers for the rapidly growing industry-wide momentum of virtual appliances.

What is a virtual appliance?

A virtual appliance is:

• a fully pre-installed application and OS environment
• that runs on any standard x86 desktop or server
• in a self-contained, isolated environment known as a virtual machine.

With virtual appliances, users can:

• Get new services up and running immediately by eliminating the need for application and OS installation and configuration
• Deploy new services easily on any existing x86 hardware with full independence from operating systems or hardware
• Better leverage IT resources by running multiple environments on a single server or desktop machine

Martin Banks talks about them in IT Week. Steve Herrod discusses them as well.

Dan Kusnetzky, executive vice president, Marketing Strategy for Open-Xchange and former vice president and analyst at IDC, says, "Customers deploying open source software in a virtual machine can reduce their software and hardware costs, and simultaneously enhance their agility in addressing future computing needs."

The Evolution of Virtual Appliances

Before last year, all virtual appliances were ones created by customers, who would use them to reduce deployment time and improve the manageability and utilization of their environments.

In 2005, major global software vendors like IBM Software, Oracle, BEA, MySQL, Red Hat, and Novell started deploying their software in virtual appliances. This allowed them to distribute their new technologies more easily so that users could try them out and develop and test with them.

In recent months, the momentum of virtual appliances has shifted to the broader developer and open source communities. Today, over sixty software vendors and open source projects publicly and freely distribute their software in virtual appliances. They can dramatically streamline the process to deploy and adopt their software, and they can instantly make their software available to the worldwide base of 4 million+ VMware users and 20,000+ VMware enterprise server customers.

Many of these virtual appliances can be found at VMTN's Virtual Appliances.

In recent weeks, I've seen a groundswell of virtual appliances from all sectors and solution areas. Some of my favorites:

• Zeus Extensible Traffic Manager
• Open-Xchange collaboration suite
• Zimbra collaboration suite
• VMware Browser
• IBM DB2 Universal Database Express Edition
• OpenEMR physician patient care tracking

Enabling technologies: VMware Player and VMware Server

One huge factor in the landslide of momentum behind virtual appliances is the introduction of VMware Player and VMware Server, hugely popular free virtualization products for desktops and servers. Users can now freely leverage VMware's proven virtualization technology supporting 32- and 64-bit virtual machine environments, all major x86 operating systems (including Windows, Linux, Solaris, and NetWare), and optimized for ease of installation and usage.

VMware Player became generally available in December, and had well over 500,000 downloads within a month of GA. VMware Server became available as a beta in February, and has had several hundred thousand downloads already as well! These have helped create a broad interest and appetite for virtual appliances, which the open source community and software vendors across the industry have flocked to fulfill.

Alessandro Perilli talks about this at virtualization.info.

The Ultimate Virtual Appliance Challenge

Are we satisfied with the current breakneck momentum and adoption of virtual appliances?

No, not by a longshot. We're looking to dramatically increase the amount of new and differentiated technology available as virtual appliances, and the most exciting thing we're doing is the Ultimate Virtual Appliance Challenge.

We are offering $200,000 (no, your eyes aren't failing you!) in prize money to the creators of the best virtual appliances.

This is like American Idol or the World Series of Poker, but for developers!

The challenge goes for another eight weeks, so get your version of Workstation (we'll give you a free temporary license when you enter if you need one) cranking and join the fun and craziness!

What better way to get your smartest friends together, build something you will get huge bragging rights for, and bank up to $100,000 for your trouble?

Here are some thought-provoking ideas to whet your competitive appetite.

That's all I've got. Send me your coolest ideas for what virtual appliances we should be encouraging folks to build, or your thoughts on what else VMware can be doing to encourage innovation, help developers everywhere, and get virtualization to every last machine on the planet!

-- Dan Chu

posted at: 00:00 | reply to the console | permanent link

...where VMware's top leadership will share their expertise and opinions on the evolving world of virtualization. Just as the service console is the dashboard for VMware ESX Server, think of this blog, "The Console", as your dashboard for understanding how VMware is driving the virtualization revolution. Strap yourself in for an exciting ride.

posted at: 00:00 | reply to the console | permanent link