Cisco enters the VDI fray with "VXI" (Related: we have a new leader in the "most press for least actual product" race!)

Brian,

You're missing the point with VXI. Saying it is just a reference architechture would be like saying VCE Vblock is just a reference architechture for VMware products, EMC products, Cisco products and management. VXI is comprable to Vblock in that it is essentially a stack for virtual desktops - including a mix of legacy technologies but also incorporating new innovations and partnerships. And like both Vblock and NetApp FlexPod, VXI centers around the myriad Cisco UCS innovations for compute. The virtual desktop market has been adrift without a framework for enabling an effective enterprise deployment. VXI remedies this.

Personally I think this is more about Cisco entering the market as opposed to where it stands today. They are too big to ignore and I am sure we will see more from them in 2011. Overall it's good that more people will be talking about our space. In the end WE the customer will decide what fits our needs. I am sure MS has to react to this in some way....

Brian,

I confess that I am completely ignorant about the HP reference architectures, but given the lack of innovation in HP's answer to UCS with its BladeSystem Matrix, I would be surprised to see anything groundbreaking. VXI, on the other hand, incorporates the myriad innovations of UCS itself along with the convergence of UC and virtual desktops. While the higher UCS VM densities are touted by Cisco as a primary financial benefit, I suspect the UC/desktop convergence is going to be far more impactful. VXI provides the ability to consolidate overlapping expenses and eliminate separate back end infrastructures for voice and VDI. It enables desktop, voice, video, Telepresence and collaboration on a single device which is accessing a virtual machine running on a UCS in the data center.

@roidude2 - I'll be more than happy to buy several VXI systems right away. Just tell me how many kids of yours are getting ready for college.

Can I at least finish my morning coffee before I get served my daily f------ dose of Kool-aid????

For just one day, let's pretend we're not a flock of grazing sheep. Please!!!

Edgeseeker,

Perhaps you should take the time to understand what VXI truly is before you blandly dismiss it.  When Cisco UCS debuted last year, it received similar (though better reasoned) reactions from both competitors and the press who didn’t comprehend what it was.  HP called it a “giant switch”. Byte and Switch called it, “next year’s servers…shipping this year”. According to ComputerWorld, a Dell executive referred to it as a “one-size-fits-all blade server”.  Of course, in a little more than a year UCS has disrupted the data center status quo, set the industry abuzz and is displacing long-term data center competitors seemingly at will.  VXI has the potential to similarly revolutionize the virtual desktop space.

Cisco VXI is the start of something more than Virtual Desktops, it's about collaboration and it seems like it will add benefits to desktop virtualization to be more successfull.

community.citrix.com/.../Prediction+-+Next+generation+collaboration+will+drive+rapid+adoption+of+desktop+virtualization

When comparing the reference architectures for Cisco, Dell, and HP IMO I was more so impressed with the Cisco + NetApp + XenServer combo than the others.

Granted I would like to add XenServer Intellicache and XenDesktop 5 into that mix. Even though I heard initially XD5 won't support intellicache, maybe a few months from release it will.

A giant lure is what it is. Go ahead and bite.

I will respectfully disagree with this being about drinking any kool-aid.  Plain and simple this is about UCS and what UCS brings to the table for virtual desktop infrastructure compared to any other reference architecture from HP, Dell, etc.  I agree with @roidude2.  We currently are a very large HP shop and we recently spent some time with Cisco in San Jose talking a lot about UCS and walked away with a very different opinion than what we walked into the room with.  After nearly a week of heavy white boarding, discussion, and debate with product managers, engineers, and architects we walked away with the fact that UCS is really the best data center platform for enterprise scale desktop virtualization. Please note, I did not say compute platform or smb.  UCS is much more than just processor and memory on a system in a blade enclosure which is really what HP or anyone has brought to the table.  It is about truly virtualizing compute, network, and storage at the frame or rack level and removing management and overhead aspects of that from the hypervisor.  Where it is also extremely relevant for desktop virtualization is the amounts of memory that put into a two socket box for a fraction of what any other vendor can do.  Whether someone is talking about dedicated or shared virtual desktop infrastructure, this is a wrapped solution that is intended to scale for the enterprise.  Just as an example, you can talk about network virtualization.  Today with any other solution, the hypervisor has to address the overhead and management of breaking up physical NICs into virtual nics for the virtual hosts in the system.  UCS handles that at the frame and not at the compute later so that the hypervisor is focused on addressing exactly what its intended for.  This may not be a huge deal in normal server virtualization, but we all know in the desktop world anything taken by the hypervisor means less resources for the virtual machine which for us is a customer’s desktop.  For desktop virtualization to work at scale is needs to provide the exact same experience for the same or less cost.  Like @appdetective what I’m most excited about is this creates attention to the space and hopefully will drive more focused innovation within the datacenter to continue to look for ways to remove what is the biggest barrier to large scale adoption, cost.

@brian

The post is trying to get people thinking that Desktop Virtualization is more than just 'better management of IT services'. It enables a people centric approach rather than the archaic device centric method which is liberating for everyone.

Any services that IT can deliver ontop of desktop virtualization will further cement it into our society. Collaborative services are no different.

Personally, I am not impressed with VXI as it is right now, but I will keep an eye on this space.

Some more interesting news, Cisco UCS + VMware View partnership announced:

newsroom.cisco.com/.../prod_111610.html

It's all about UCS.

@tjbruni13 - Wait until HP pays you a visit to articulate their own vision. I bet you'll walk away with yet another outlook on things.

Look, it's plain and simple. This site has done a great job over the years articulating VDI's  roadblocks, shortcomings, and its state of readiness. How does VXI help overcome these challenges? It's a marketing ploy, that's all it is.

@edgeseeker - Met with HP about a month before Cisco.  Saw the long term vision.  Interesting story around strategies on power and cooling controls.  However, nothing to the level of CIsco strategy to truly merge compute, network, and storage into central framework where infrastructure growth just adds on instead of adding individual components.  Plus HPs VDI specific infrastructure was about putting compute and storage in a box/rack which is only about 2/3 the way there.  Like I said before I think Cisco is going to force innovation and drive down costs to compete which is good for all of us.  Also, I know Brian joked about the cost being equivalent to putting kids through college.  This is probably true when you compare for small scale (couple hundred users).  If you are talking about enterprise scale (thousands of users) which is row or rows of servers depending on the size of your data center, Cisco will be a lot cheaper than the competition.

@tjbruni13 - I believe you. Let's see how things develop moving forward.

@appdetective - I hear you, but I think they're too big to take seriously, frankly. Cisco wants to have a dog in every fight, and we know where that almost always leads: to the dog house.

@edgeseeker

nice analogy.

reminds me of Oracle VDI...

Looking at the announcements yesterday, I felt the most important part was the upcoming thin+phone devices.  Especially since Cisco is working with the existing VDI vendors on the back-side.  That makes sense to me.

What would far more interesting than a phone with bolted on thin client, would be to see Cisco integrate a SOC version of thier UCM SIP Softphone onto a regular thin client.

That way, we could have reliable, task worker IP telephony in with a headset with minimal space impact.

At the moment, I see callcentres and offices full of task works with seperate digital telephony systems that cost almost as much as thier RDS + thin client environments. Due to the varous limitations on audio streaming in most RDS setups (VDI is a bit different, but still limited) they cant integrate these systems to reduce the amount of hardware investment (and management overhead) on a desk.

So how's about a regular thin client box (with options for rdp/pcoip/hdx/remotefx) with some built in SIP circuitry and a plug to stick a headset into - run the SIP on the same POE cable a seperate VLAN (Like the LAN passthru mode on the Cisco deskphones do already) and they've got an instant winner of a hardware product on thier hands.

The VXI "platform" doesnt really excite me at the moment - it seems like a good way for Cisco to crossmarket thier existing technology stack into the emerging Virtual Desktop market, without bringing anything to the table that isnt already available (and possibly a lot cheaper and more flexibly) by a variety of other vertical providers.