We recently wrote a piece for VMBlog's series predicting what lies ahead in 2011 for Desktop virtualization.  


Here's an excerpt of the article - what do you think? Are we on the right track?

2011 - Predictions of what lies ahead in Desktop Virtualization

The year 2011 is going to be very exciting for desktop virtualization.  Why? Because several trends that have been building steam in the last couple of years are joining forces to create a big splash.  While no one knows exactly what the future holds, here are my predictions, derived directly and indirectly from numerous real user experiences, for the upcoming year:

• Mobile devices become the perfect desktop clients. I'm sure this one comes as a surprise to no one. More companies will adopt mobile devices such as the Apple iPad or Android aPads as the primary ways to access their corporate desktops. The growing popularity of these devices is driven by end-user demand. From executives to students and sales agents, users appreciate the anytime, anywhere access these portable devices provide.
  
  For IT, this means finding an efficient way to deliver and support the corporate desktop (typically Windows based) on these devices without having to take on the added support burden of managing a wide array of mobile devices.


• Virtualization becomes an embedded feature, not a specialized competency. As the use of virtualization grows across every aspect of IT, it is becoming an embedded feature of products rather than a core competency of its own. We are seeing virtualized networks, servers, desktops, storage, and applications. As this trend grows, it is untenable to view virtualization as a specialized IT competency that is managed by a handful of datacenter/virtualization experts. In 2011, we will see vendors increasingly embed and simplify virtualization within their products so that IT of varying skill sets can easily setup and manage the solutions themselves without needing specific virtualization expertise.


• Cloud economics permeate in-house virtualization. Cloud computing offers three compelling benefits that customers find attractive: elasticity (grow or shrink capacity on-demand), pay-as-you-go, and the convenience of a managed service. Since these benefits are primarily derived from multi-tenanting/sharing a physical infrastructure across multiple customers, the same results have not been achievable with in-house virtualization. This is changing especially as virtualization vendors are paring down the infrastructure needed to run their solutions. This year we will start seeing similar economic benefits with in-house (private) clouds as we see with public clouds -- particularly in specific application areas such as desktop virtualization and storage virtualization.


• Rise in integrated end-to-end solutions. Why assemble your own car if you can buy one for less? In many ways, we are still at the "assemble-it-yourself" step with virtualization and cloud computing. While the market is still too early to see widespread consolidation, we will see more integrated end-to-end solutions from best-of-breed products. For instance, we will see more turnkey appliances that are pre-configured with the hardware, virtualization stack, and applications.
  


• Demonstrable capital expense reduction with Virtual Desktops. Historically, the business case for virtual desktops (VDI) has been based on long-term soft dollar operational savings in return for higher upfront capital expense. This has been a tough pill to swallow for many, and cost and complexity are two key hurdles to widespread adoption. In 2011, we will see more examples of demonstrable capital savings from VDI deployments based on second generation VDI architectures that require significantly lower capital investments than traditional PCs.
  


• VDI production deployments accelerate. For a few years now, analysts have speculated as to when the VDI market will truly hit its stride. 2011 will be the year we see a record number of VDI deployments in production - this is based both on the success that VDI pilots are enjoying this year, and based on the newer VDI architectures that address the cost and complexity barriers to adoption.

In a nutshell, I see 2011 as the beginning of a multi-year growth in production deployments of virtualization and cloud computing that result in greater access, better uptime, at significantly lower costs. 
About the Author
Krishna Subramanian leads marketing and business development for Kaviza. She has 16+ years of software industry experience with expertise in cloud computing, virtualization and Software-as-a-Service. Prior to Kaviza, Krishna led business strategy and acquisitions for Sun Microsystems' cloud computing and software businesses that delivered over half a billion dollars of incremental revenues. Before Sun, Krishna was the CEO and co-founder of Kovair, a Software-as-a-Service CRM company that she ran for 4 years and grew it to a Computerworld Top 100 Emerging company with major enterprise customers. Earlier in her career, Krishna had worked at Sun Microsystems for over 6 years in various roles managing Java software products. Krishna received YWCA's achievement award, and has been interviewed by San Jose Mercury News, Red Herring, Business Week and others. Krishna received her Masters in Computer Science from the University of Illinois, Urbana-Champaign.

There is a debate raging on BrianMadden.com and elsewhere as to whether local storage can be used with VDI, and if it is good enough.

This strikes at the core of why we founded Kaviza, so here are our perspectives. Folks are debating between 2 options:

Option 1: Shared storage. You get high availability, dynamic load balancing, etc but cost is high.

Option 2: Local storage with traditional VDI. None of the above but cheap. 

They are not talking about Option 3. Where is Option 3??

The conclusion here seems to be that Option 2 is good enough for VDI.  Or is it? What are you losing? The main reasons people go to VDI is to streamline management, and increase desktop uptime.  Will Option 2 give these?

i) What happens when you go beyond 1 server? Do you manage all the desktop images manually on each server? Keep them up-to-date? Statically provision to each server?

ii) What happens when a server fails? These are not server workloads, but users care about their desktops being up. No redundancy means downtime.

The critical question is whether we can retain the inexpensive, high performance, and obvious choice of local storage in VDI when we need scalability and high availability.  Can we get a scalabile, highly available desktop service without the high upfront and ongoing costs of shared storage?

So, what if you had an Option 3?

Option 3: Local storage with a new VDI architecture designed to deliver high availability and dynamic load balancing on DAS.

This can be done - in fact, it's exactly what Kaviza has.  Again, this is not aimed to be a plug but to technically highlight how an architecture can achieve DAS and HA.

Kaviza was founded on the notion that those features critical to VDI can be retained in a distributed, local-storage architecture by implementing an abstract common store of the images through a desktop-specific management layer.

Kaviza maintains the desktop images locally on each server.  The actual desktops running on each server, generally as thinly cloned VMs, use the local images solely.  As admins go through image lifecycles to update and patch the base images, those changes are automatically transferred to the local storage on the other servers that need them.

Thus the appearance to the administrator is that there is a single store common to the entire installation.  But the implementation of the store is entirely on local disk.

As new servers are added to the installation, those images that need to be sent to them are automatically transferred there.  And if a server is lost due to failure, the desktops running on that server are automatically reprovisioned on surviving servers.  Hence Kaviza supplies both scalability and high availabilty using only inexpensive local storage.

For many customers, desktop forklift to justify high investment in traditional VDI is a non-starter. What they want is a solution that gives them an immediate return on their investment for every refresh cycle, no matter if its 30 desktops or 3000.  For these customers, local storage is the only way to go.  And Kaviza has a comprehensive virtual desktop solution that linearly scales throughout that range at a constant price per desktop.

Labels: Brian Madden, DAS, desktop virtualization. VDI cost, Kaviza, Kaviza Citrix HDX virtual desktops user experience performance cost affordable low-cost price PC refresh Windows 7, SAN, Shared storage

Today we announced that Kaviza has obtained a license to resell the Citrix HDX technologies.  With this, Kaviza now addresses all three roadblocks to virtual desktops: cost, complexity, and user experience.


The upcoming Version 3.0 release, which will be available in beta by the end of May, will include an add-on to run virtual desktops with Citrix HDX. 

We are very excited to make this announcement because we now offer unparalleled value to customers - simple, affordable, turnkey virtual desktops with rich optimized user experience for less than the cost of a PC! No one else can provide this.  Virtual desktops are now a truly viable alternative to PCs - Kaviza makes virtual desktops less expensive to buy, less expensive to operate and same or better performance as PCs.

How can you say no to that?  Let us know if you would like to try the Version 3.0 Beta.

Labels: Kaviza Citrix HDX virtual desktops user experience performance cost affordable low-cost price PC refresh Windows 7

Its been crazy busy, so I haven't had time to blog in several weeks!  Good things are happening! 

Gartner picked Kaviza as a Cool Vendor in Client Computing for 2010.  Yes!!

From our briefings with Mark, it is clear that Gartner understands the need for a simple, affordable virtual desktop solution that can be deployed within desktop refresh budgets by desktop IT.  Apparently they get a huge amount of calls on desktop virtualization, and cost and complexity are two key issues customers bring up.

This is consistent with our experience.  Simple, quick, affordable, open, flexible - these are the primary points we strive for.

Speaking of which, another company that is doing interesting things around making IT efficient, easy-to-procure, and easy-to-manage is Matrix42, who is leading the charge with "IT-Commerce". They are a strong market leader in Germany and are growing operations worldwide.  We announced yesterday that we have entered into a partnership with Matrix42.  They will co-brand and resell Kaviza VDI-in-a-box as the preferred virtual desktop solution that users can procure from their IT Service Catalog.  This is very exciting, and we look forward to a strong collaboration with Matrix42. Checkout the Kaviza microsite on their site.

We are heads-down preparing for the Citrix Synergy show May 11-14th in San Francisco, CA, more on that in the next post!

Labels: affordable, award, Citrx Synergy, Cool Vendor, Desktop Virtualization, desktop virtualization. VDI cost, Gartner, Kaviza, Matrix42, virtual desktops

We announced last week that we took an investment from Citrix. You can read about the investment in our press release and the media articles that covered it. 

Why did we choose this investment?  Quite simply, because we felt this would help us best serve our goals of  leading the market in making desktop virtualization affordable and simple. 

The investment by Citrix validates Kaviza's business model and architectural design.  Our grid architecture fundamentally changes VDI economics by eliminating the expensive infrastructure costs that make current approaches tough to digest.   

We are also excited by the opportunity to work together with Citrix going forward.  We are going to be at the Citrix Synergy show in May, so if you are planning to come, do stop by booth #613.

There is a lot of speculation about whether 2010 will be the year VDI hits its stride.  We all know intuitively there is a huge market for desktop virtualization given the large installed base of PCs that need to be refreshed - the questions are, how quickly will this market adopt desktop virtualization and is the technology addressing needs sufficiently  to drive broad market acceptance?

Our point-of-view, from having worked with VDI over the past decade, is that the demand for virtual desktops is real, but for widespread adoption two things are needed: the technology needs to be less expensive than PCs and it must be easy to use.

Interestingly, several studies have pointed out the same - just this week I just saw an article from a business consulting firm who said that most VDI proof-of-concepts are failing.

The first two of their three reasons relate to cost and the corresponding value.  As they say, "customers must replace cheap desktop computing and storage with expensive datacenter computing cycles.  This is a major upfront capital expenditure that is difficult to justify in a tough economy."

We consistently hear this from our customers - companies want to do more with less. The question they ask is: "How can I use my PC replacement budget to deploy virtual desktops and get a lower TCO without needing a datacenter forklift or changes to my business processes?"

This is the exact problem Kaviza addresses - our patent-pending architecture eliminates the need for expensive infrastructure and shared storage, and scales on inexpensive commodity hardware.  All the desktop provisioning and management functionality is built-in, so you can simply load Kaviza on any hypervisor-enabled workstation grade server, and go, and it is designed to be managed by desktop IT.  Try how easy it is, and let us know what you think.

Labels: complexity, cost, Desktop Virtualization, fail, hot trend, Kaviza, proof-of-concept, Terminal Services, VDI

A news item caught my attention a few days back:  a Pa. school district was accused of using the webcams on school issued laptops to spy on a student at his home.  This raises an interesting issue of where to draw the line between access and privacy - especially as several school districts are considering issuing laptops to middle and high school students.

The reason schools want to provide students with computers is benign and noble - they want students to easily have access to all the software programs and tools they need to do their school work without having to buy and install these themselves. But the question is, should schools be issuing laptops?  Not only does this raise privacy issues, but it also adds to management and oversight headaches - schools now have to keep these laptops patched and up-to-date, they have to track these assets, update them, etc.

There is of course a better way - school districts could provide students with virtual desktops that have all the software applications and tools the student would need to do their school work.  Students can access these virtual desktops from anywhere - from their home computer, from a computer at the library, etc.  This solves several issues: i) schools don't have to be in the business of managing physical laptops or patching them - they simply update the templates for these virtual desktops periodically and the changes get propagated automatically the next time the students login, ii) privacy is ensured since the virtual desktops can be locked down to restrict access, so schools can enforce policies easily using the desktop templates, and iii) this option provides students with anytime, anywhere access without having to carry around a laptop and manage it.

All the school districts we speak with are very receptive to virtual desktops - they love the simplicity, the ease of management, and the affordability that solutions like Kaviza's offer.  Don't take my word for it, try it for yourself.

Labels: Desktop Virtualization, Kaviza, privacy, School laptops, security, VDI, virtual desktops

I am always curious to see who Time Magazine picks as the Person of the Year because their pick not only honors that person's achievements but more importantly, it recognizes a key theme or trend that was a major focus for all of us that year.


I wish Time would do the same for companies - if they did, my pick for 2010 would be the Small to Medium Enterprise (SME).SME's today can enjoy the same superior technology infrastructure, sophisticated forecasting applications, and cutting edge innovations for a fraction of the costs that large enterprises paid for these innovations.

Who would have thought a few years back that you could now have a global CDN for pennies, thanks to cloud computing players like Amazon, or that for a few dollars a month, you could have sophisticated sales forecasting or marketing automation applications thanks to the Software-as-a-Service (SaaS) companies.

These "as-a-Service" and cloud  innovations with flexible pay-as-you-go models level the playing field, enabling SMEs to compete with the same level of sophistication as the big guys.  But, there is one area where this cost disruption has not yet happened - on-premise virtualization.

Virtualization infrastructure still requires massive datacenter buildouts, with expensive shared storage, and custom built setups that are complex and labor-intensive to manage.   But it doesn't have to be - especially not with desktop virtualization.  Virtualizing deskotps streamlines and drastically cuts down desktop support costs, improves uptime, and gives companies a competitive edge where it matters most - its employees' productivity.  This is of value not just to large enterprises, but perhaps even more to SME's, who have to do more with less.

Which is why Kaviza has re-designed the hosted desktop virtualization stack so companies do not require the expensive datacenter buildouts or shared storage pools to run virtual desktops - for under $500/desktop, starting with as few as 25 desktops, a company can quickly and cost-effectively get immediate ROI on virtual desktops with Kaviza.  Try it for yourself, and see how Kaviza is changing the game in desktop virtualization.

Labels: Cloud Computing, Desktop Virtualization, desktop virtualization. VDI cost, Person of the Year, SaaS, SME, Time Magazine, virtual dekstops

I was recently asked by CRN magazine to comment on whether I agreed with VMware that VDI is poised to take off in 2010.  I said yes, but for different reasons than VMware. You can read the article here.

As a server virtualization vendor, VMware sees VDI as an extension to server consolidation, which is why they believe VDI will be driven by companies that have already adopted virtualization.

But we see a much larger opportunity for VDI - it addresses a desktop problem that every desktop IT administrator is interested in solving, assuming they can make the math work.  Desktop IT wants to address the operational headaches and cost of managing PCs, but they need to solve this problem within their existing desktop budgets.  They are driving the agenda for VDI, and its not about datacenter consolidation.  With new architectures like ours making VDI less expensive to acquire than PCs, desktop IT can now use virtual desktops as an alternative to refreshing PCs, an option that didn't exist a year ago.

We believe this is the driving force that will make VDI attainable for everyone, and will drive an upswell in adoption.  We are definitely seeing this, and we look forward to an exciting 2010!

Brian Madden had an interesting post yesterday on why a non-technical business person would be interested in VDI.  Brian surmised "the same benefits of Terminal Services without the hassles" as one compelling reason. This makes sense, because with VDI you don't have to learn a new OS(Terminal Services), you don't have app compatibility issues, and you can run the full desktop.

But, does traditional VDI bring with it a new set of hassles bigger than the ones it replaces?  We talk to several customers who want to upgrade from Terminal Services to VDI, but are daunted by the cost and complexity of traditional VDI - the ROI is unclear, traditional VDI requires a cultural change away from the desktop team as it involves managing memory pools, virtualization resources, and the top-down VDI model doesn't fit how desktops are typically upgraded (in phases).

What  our customers tell us is that VDI as an upgrade to Terminal Services is compelling IF:

• The math works: If they can get VDI at the cost of a PC, without a huge upfront investment.
• No cultural change is needed: Desktop IT should be able to deploy and manage the VDI solution
• Companies don't have to change how they buy/upgrade desktops: which typically happens in phases (e.g. as desktops age, or one department/branch at a time). Traditional VDI is typically not cost-effective below a few thousand desktops because of the massive upfront investment it requires.
  

This is precisely what Kaviza delivers - all the benefits of Terminal Services, without the hassles, and with no new costs or complexity.  Try it for yourself.

Folks,

 Happy New Year! We hope 2010 is a joyous and prosperous year for everyone!

The amazing thing in our global economy is how truly "around the clock" things are these days.  To give you an example, virtualization guru Doug Brown posted a videocast on Kaviza and a podcast interview, both on December 21, 2009.  As a VDI expert, Doug completely "gets" the value of simplicity and affordability - I had a wonderful time recording these sessions with him, and we had a long conversation offline chatting about the space afterwards.

Anyway, expecting things to be slow during the holidays, I thought I would wait until this week to blog about it.  As it turns out, we received a bunch of downloads on our site the last two weeks from all around the world.  I guess the quiet holiday time is a good occasion for folks to try out new products!

Speaking of which, we just released our latest version with support for Windows 7, a gateway for secure SSL-based remote access, and a standalone Java client.  Try it out if you get a chance, and tell us what you think!

Computerworld wrote an interesting article on 5 shortcomings they see with VMWare View4 - one of which is its high bandwidth requirements over WANs.  The article quotes Chris Wolf from Burton Group as saying Kaviza's VDI appliance approach overcomes this.  The article goes on to describe other shortcomings, including cost - although VMWare claims View4 can cost as little as $750/desktop, they concede this is assuming you deploy at least 2,048 users, and use the VMWare-Cisco-EMC storage-server-software package AND use a repurposed PC.  This means a minimum investment of $1.5M is needed to get such a price advantage, and you are still stuck with old PCs!

Kaviza, of course, delivers virtual desktops at a total cost (including hardware, software, storage) for under $500/PC, and this is true for both small (as little as 25 desktops) to very large (2,000+ desktops) deployments.  How does Kaviza accomplish this?

Kaviza is the only VDI vendor that has rearchitected the entire stack above the hypervisor to be focused on desktop virtualization, and hence runs on inexpensive commodity servers with direct-attached storage.  Kaviza lowers the infrastructure costs of VDI by about 1/3rd to 1/4th by eliminating the need for shared storage and multiple servers. 

Furthermore, as Chris points out, Kaviza offers total flexibility in deployment - you can place Kaviza VDI appliances at remote/branch offices and manage them centrally for better performance in bandwidth-constrained environments or co-locate all the servers centrally.  Kaviza provides this flexibility because Kaviza's Grid architecture is inherently distributed - it does not rely on centralized shared storage, thereby allowing the various appliances on a Grid to be distributed at different locations.

See how easy it is to setup and manage a Kaviza grid, download the free trial today!

See how easy it is to setup a Kaviza VDI-in-a-box^TM server and start deploying virtual desktops! Checkout the demo, and download here to try it for yourself.

VMWare's executive recently talked about how he believes Windows 7 will be a catalyst for desktop virtualization, but expects adoption to be slow. What he doesn't tell you is an important reason why: cost. If its going to cost you $1000-2000/desktop to upgrade from XP to Windows 7 either with traditional PCs or virtual desktops, CIOs are asking if its worth it.

Gartner's VP Research, Michael Silver estimates that when you include replacement hardware, admin costs, application testing, and replacing incompatible apps, -- in a hypothetical organization with 2,500 Windows users -- the cost of upgrading from Windows XP to Windows 7 will run $1,035 to $1,930 per user. For 2,500 users, this is a $2.5M-$5M effort, which is substantial. Migrating to first generation VDI architectures such as View would be no cheaper. Analysts estimate the average upfront cost/desktop to be $1,200-2,000 for traditional VDI.

But what if you could upgrade from XP to Windows 7 for under $500/desktop, and get all the benefits of virtual desktops, with lower operating costs, greater security and the option to preserve the XP desktops to smoothen migration woes?

We are announcing today Windows 7 support with Kaviza's VDI-in-a-box^TM , a beta is coming in the next few weeks. You can get a Windows 7 virtual desktop for under $500, and this includes all the hardware, software licensing and storage costs. We are already getting a lot of partner and customer interest in this, so its going to be exciting!

Click here for a free trial of Kaviza's VDI-in-a-box^TM.

Labels: desktop virtualization. VDI cost, PC upgrade, VMWare, Windows 7, Windows 7 cost

Here is a video chalk talk describing why we founded Kaviza, the shortcomings of VDI today and how Kaviza's architecture addresses VDI cost & complexity. Take a look and let me know what you think!

Labels: Citrix, Desktop Virtualization, Kaviza, VDI, Video, VMWare, YouTube

Last week, UK's, "The Register" conducted a survey on VDI in which they found that although customer interst in VDI is high, customers see cost as the primary roadblock to adoption. This exactly matches our experience. We founded Kaviza to eliminate the cost and complexity of VDI. Here are some highlights from the survey resuts:

Server-based VDI was second only to Terminal Services (TS) as the most adopted form of desktop virtualization. Given how long TS has been around and how new VDI is this is quite remarkable and it validates the sort of interest that we are seeing among both customers and partners.


So if there is so much interest and understanding why aren't deployments going through the roof? What's holding VDI adoption back? Well, the latter questions of the survey bring that out well. It has to do with the costs.

Cost was picked by respondents as the major reason for not moving forward with VDI. Over 75% felt the upfront infrastructure and implementation costs of VDI was the biggest roadblock, followed closely by the high storage costs.

Of course, this is exactly why we founded Kaviza two years back. Having led VDI for large companies, we saw how expensive it is to implement traditional VDI. This is because traditional VDI is layered on top of a generic server virtualization platform and uses the same heavy infrastructure that was designed to virtualize a few mission critical applications (in the hundreds) as opposed to thousands of desktops. And therein lies the issue. Current VDI approaches require big-iron servers, shared storage like SANs, high-speed interconnects - all of which are expensive to acquire and deploy.

At Kaviza, we realized the only way to bring down the cost was to start from scratch. We set out to design a VDI architecture that could run on inexpensive commodity whiteboxes and scale horizontally like modern Web architectures (e.g. Google). No more high-speed interconnects and shared storage and servers with expensive FC HBA cards. Kaviza's VDI-in-a-box is a distributed next gen VDI solution that provides virtual desktops at sub-$500 prices (the pricepoint is the fully loaded cost of a virtual desktop, including Microsoft's VECD license and all the infrastructure). We recently launched a Beta, and the response has been overwhelming.

If you'd like to try it for yourself, check out our free trial. In and hour or two you should be up and running with a single server and many desktops. If you want HA, then email us to get the license that will allow you to tie many such boxes together without the need for shared storage, high-speed interconnects and all the other stuff needed for mission critical server virtualization. In case you want to read the articles: click here to participate in the survey, its still open. See the results here.



Kumar

Labels: Desktop Virtualization, poll, Register, VDI cost, VDI survey, virtualization

Today is an exciting day for us at Kaviza. We launched a beta  of our multi-box virtual desktop appliance which offers built-in high availability. 

You can setup a highly available farm of virtual desktop servers by simply loading Kaviza on two or more commodity servers and linking them together.  It is truly that simple - no custom coding, no expensive SAN's, high speed interconnects, or high end servers are needed.  Our entire solution can be setup from scratch in under three hours.  With Kaviza's unique shared-nothing, Google-esque architecture, you get a highly scalable solution at a very low cost.  There is no single point of failure, and features like high-availability are built-in.

And the best part is that our virtual desktops cost less than a PC to deploy - our fully loaded cost (including the server, our software, MS, the hypervisor, and other licenses, thin client) is cheaper than a PC.

We are encouraged by the response we are getting from early trial users on this release. Here is some feedback we have received from beta partners and customers:

"Now I see why you say it's "turnkey". We got the whole thing up and running on multiple servers in just a few hours."

"Neat architecture, solves the fundamental awkardness with VDI."

"My team is raving about your solution."

You can either download a free trial of our single box version here, or register here if you would like to try the multi-box version.  We would love to get your feedback, let us know what you think!

Labels: Desktop Virtualization, High-Avaialbility, New Release, Product Announcement, VDI

Labels: Brian Madden, desktop virtualization. VDI cost, Kaviza