Citrix's ICA problem is still a problem for widespread VDI adoption

Citrix's release of XenDesktop (and VMware's crazy response) are still dominating my thoughts this week. Back in January, I wrote that Citrix XenDesktop would be a stronger product than VMware's VDI offering.

Citrix's release of XenDesktop (and VMware's crazy response) are still dominating my thoughts this week. Back in January, I wrote that Citrix XenDesktop would be a stronger product than VMware's VDI offering. Now that XenDesktop is real, I still prefer XenDesktop over VMware VDI (and for the same reasons I did in January). But instead of saying XenDesktop is stronger, perhaps it's more appropriate to say that XenDesktop is the less incomplete of the two?

Many of you know that I love the concept VDI and that I think VDI will ultimately replace TS-based desktop virtualization in a few years. (In fact, I'm taking the "VDI" side in the "VDI versus Terminal Services" debate at BriForum in a few weeks.) But I want to point out that I love concept of VDI if it's done right. And today, neither Citrix nor VMware do it right. (For the record, I am 100% convinced that both Citrix and VMware will do it right after a few more product revs—maybe after another 18 months or so. But today VDI is largely a hypothetical mental exercise for me.)

The fact that today's remoting protocols have some limitations seems lost on VMware and Citrix--at least in their marketing messaging. For example, the only "real limitations" of XenDesktop that Citrix's Sumit Dhawan mentions are "offline access requirements or advanced peripheral support." In last week's letter to partners, VMware's Jeff Jennings wrote that "One of the main value propositions of a virtual desktop is that all your applications work in a VDI environment." Neither vendor mentions the real inconvenient truth—that neither ICA nor RDP can remote all applications.

How important is it that the remote display protocol does "everything"?

Citrix's ICA protocol is light years ahead of Microsoft's RDP protocol, which is the protocol VMware uses for their VDI solution. (Unless you buy specific thin client devices—err, "desktop appliances"—that include other protocols, like Wyse TCX.) In fact I think that ICA is a huge reason to use XenDesktop over VMware VDI. But like I said, ICA still can't do everything.

How close is ICA to doing "everything?" 90 percent? 92? 99?

Actually it doesn't matter. Because anything less than 100.00% means there will be some use cases where ICA (and therefore XenDesktop) doesn't work. And if this is the case, then a customer will need some method in addition to ICA to deliver 100% of their apps. And if they have to introduce an additional method, who cares whether they have to use that additional method for 1% or 5% or 50% of their apps? Once they break that "additional method" barrier, they've just added huge complexity to their environment.

Think of it like this. If a customer cannot remote ALL of their apps, then they have to start start making choices about which apps get remoted. They start going through their app list one-by-one, saying yes, yes, yes, no, no, yes, no, yes... They ultimately end up with a mixed solution of some remote and some local apps.

A mixed solution of some remote apps and some local apps.... Sound familiar? It should, because that's exactly what everyone is doing today. Some apps are installed or streamed locally, and some are delivered remotely via TS-based solutions. This is fine, and widely accepted.

What's the value of partial VDI?

Most people would probably agree that the concept of VDI would be cool if it would work for 100% of desktops, users, and apps. Everyone agrees that we're not there yet—at least not with Citrix or VMware. So "full" VDI is a future concept.

If today's environments are a mix of local and remote, and if today's VDI only works in certain use cases, why even bother introducing VDI into an environment if it can't get rid of the TS-based apps and the local apps? Now you'd just have three different app architectures instead of two. (With the new architecture introducing a whole new set of servers, products, costs, and complexities.)

Citrix's Sumit Dhawan blogged about the value that XenDesktop can add to an existing XenApp environment. Quoting directly from his blog, Sumit listed three three benefits of using XenApp to deliver apps into XenDesktop:

  1. Dynamic provisioning of virtual desktop implies that a user's desktop always stays pristine with no apps installed - all apps are delivered (using streaming or hosting technologies) enabling an on demand assembly of personalized desktop at the time when a user logs on.
  2. Predictability and Capacity planning on VDI - Separating all LOB apps that have unpredictable (problematic) resource requirements, and running them on separate XenApp servers, prevents over-provisioning the VDI server architecture and can reduce the number of servers required for virtual desktops, improving the TCO of virtual desktops.
  3. Application and license management - each app can be controlled granularly. You have complete visibility into who has access to the applications and who accessed which application when.

I don't fully agree with any of these points. To each, I respond:

  1. True, but if you're using XenApp for a shared (stateless) desktop, then why even bother with XenDesktop? Why not just serve desktops from XenApp?
  2. True, but (again), why even bother with XenDesktop? Why not just build a XenApp "desktop silo," and serve the desktops in a much most cost effective way from a TS-based XenApp server that double-hops to XenApp published apps?
  3. True, but (again still), why add XenDesktop to this mix?

I agree 100% with Sumit's three listed advantages listed above. However, these are advantages of pure XenApp environments with app silos plus desktop silos. These are NOT advantages of combining XenApp with XenDesktop. In fact, this points to disadvantages of XenDesktop, because it just adds extra cost and extra complexity into an environment.

So for the general desktop user, why even bother? Sure, much like Terminal Server, there are very specific, very niche scenarios where VDI makes sense today. And for these cases, great! Go for it! Use VDI. But for the general user population, the industry needs a display protocol that works for 100.00% of applications. If even 1% of the apps don't work, a customer would have to build out some additional solution to deliver those apps. And if that's the case, then why even bother with VDI? (By the way, Qumranet and Teradici get a lot closer to 100.00% remote display protocol app compatibility today. I'll write more on those two solutions next week.)

Citrix's other ICA problem: ICA on XenDesktop is NOT the same ICA that's on XenApp

I wrote about the implementation of ICA in Citrix's XenDesktop product called PortICA a few months ago. Now that XenDesktop is out, Citrix has confirmed that several of the ICA features in XenApp are not available in the implementation of ICA in XenDesktop. And they've done it in the typical Citrix way—they've hidden and spun the crap out of it. Check out the quoted text from the official XenDesktop Technical FAQ:

Q: How do the ICA capabilities in XenDesktop compare with that in XenApp?
A: All of the ICA functionality of XenApp 4.5 FP1 is available in XenDesktop. The following are not yet supported:

  • Kerberos SSPI or SmartCard Virtual Channels
  • SpeedScreen multimedia acceleration & zero latency
  • PDA sync, TWAIN, shadowing and SmartAuditor
  • Audio on Vista
  • ICA perfmon counters (SMC) and end-user experience metrics

I love how they say "all" of the functionality is supported, but then go on to list a bunch of thing that are not supported. :) I also love how they have five bullet items, but they put stuff like shadowing and PDA sync together on the same line to pretend like it's only five things that aren't supported, when really it's more like eleven. Sigh!

I'm sure ICA will get there. Maybe Citrix will license something like Qumranet's Spice protocol or Teradici's PC-over-IP (and call it ICA-plus)? Maybe VMware will license something? But until that time, XenDesktop and VMware VDI are niche solutions only.

Join the conversation


Send me notifications when other members comment.

Please create a username to comment.

Agreed on all points Brian, we've had pretty good luck with some of our apps going into the PS/TS environment, but we just don't have the server space or manpower to support yet another architecture.  I'm sure like most customers, we're going to wait on the sidelines until someone gets to 100%.
Well said!

If XenApp supports those features, why are they not in XenDesktop? Has anyone heard an official response about those missing features? Which will be added in the future and which never will?

Give me TWAIN, PDA, SpeedScreen for Multimedia and Flash, shadowing and smartauditor and I am happy.

At this rate, even slow moving Provision Networks will achieve all of these features before VMware or Citrix. I wish I could hibernate for a few years and wake up when the dust settles :)
The dust is not going to settle anytime soon, "The future of Windows is heterogeneous, The future of Windows is heterogeneous, The future of Windows is heterogeneous."

From my perspective, after talking to hundreds of customers about Desktop Virtualization, they are okay with a solution that doesn't apply to every client.  Sure they would like a solution like that, but if they can reduce the desktop management costs for even 50-75% of users, that's better than what they're getting with no centralized application delivery mechanism, i.e. TS or Citrix.  Additionally, there is a common theme that TS & Citrix is just too complex, and not enough of the IT Staff can manage it.  Most enterprises have lots of IT people, and only a select few that can deal with TS & Citrix without breaking it. For companies that already have a virtualization infrastructure in place, virtualizing "some" of the desktops reduces their management costs, empowers their desktop support engineers and provides a consistent desktop regardless of location (assuming there is connectivity).

I agree that not being able to serve 100% of apps via remote display protocol is not panacea, but what most companies have now can be dramatically improved my desktop virtualization, even if it doesn't fit every use case.

As for the "slow moving Provision" comment, I'm not sure what that means, as we seem to be able to add features at a more rapid pace than the competition.  We typically have two releases every year (not service packs), and each packs more and more functionality on top of a management framework that has surpassed what Citrix has offered for the past 8 years.  What exactly do I mean by that?  Since 2000 we have been delivering tools that "enhance" Citrix and Terminal Services to provide features like:

End-to-end Universal Printing

Granular Hybrid User Profile Management

Complete Desktop Configuration & Lockdown

Application & Host Access Restriction

Network Drive & Universal Network Printer Mapping (w/o scripting)

File, Folder & Registry Redirection to make applications work that typically required administrative rights/permissions

Add/Delete registry keys or values at logon, without any scripting.

Assignment of apps/desktops to Users, Groups, OU, Client IP Address/range or Client name.

Full-featured CPU & Memory Management, without springing for XenApp Enterprise.

Integration with Softgrid, so softgrid apps can be published and accessed seamlessly or via published or virtualized desktop.

Fewer clicks to publish/unpublish multiple applications, i.e. unpublishing apps from a Terminal Server or Managed Desktop Group is two clicks.

One single integrated management console and configuration database that manages everything, i.e. Terminal Services, Virtual Desktops, Physical Desktops, Desktop configuration & lockdown, profile management, policies, printing, softgrid....

Oh, did I mention that one doesn't need XD Platinum edition to shadow Virtual Desktops.  I wrote the spec for this and it's being added to our product in June.

As for RDP, we already extend RDP to offer true Multi-monitor support, Windows Media Redirection, local text echo, bi-directional audio, usb handheld sync and have a prototype of speedscreen for RDP.

I'd say that we move with more agility than "you know who", and have fewer holes to fill.

The next few years are going to be exciting.  It'll be interesting to see what becomes of the Net2Display specification, or if one protocol becomes the industry standard.  It's cool that Citrix has ICA,  and that we're extending RDP, but it'd be nice to have a newly designed protocol that doesn't have any legacy baggage.

My 2 cents.

Additionally, I had someone tell me that XenDesktop is Hypervisor Agnostic, as I had been using that terminology for our Virtual Access Suite.  I have since stopped using that term, and instead say "tight integration and automation", since supporting a hypervisor is more than just power management.  I think at a bare minimum one should be able to power manage, clone and delete virtual machines on a given hypervisor to say that there is any real integration.  Pointing one console to another console and importing VMs isn't too integrated.

I think the main brake in the deploiement of VDI is the cost :

-> Powerfull server for just 20 users instead of 100 users / TS,

-> the licence cost (client OS, VDI licence, server licence),

-> SAN Storage and backup,


A 2x4 Core system with sufficient RAM (16-64GB, depending on how much one allocates to each VM) often can scale much higher than 20 users, i.e. 30-50.  Parallels Virtuozzo can scale to over 100 concurrent VDI users on a single server. It's not the same kind of desktop virtualization, but an option some of our customers are excercising.

Some customers already maintain client OS licensing, Terminal Server CALs and sometimes Citrix Licensing.  That Citrix License costs more than the OS license, so one can do VDI instead of Citrix w/o incurring more licensing costs, assuming the same number of users.  Customers w/o any Terminal Services or Citrix infrastructure have the biggest challenge.

SAN Storage is not inexpensive, but it also is not so expensive that it's prohibitive for even SMB customers.  For 100 user VDI Deployment w/ no fancy thin provisioning Snap or Flex Clones, about 400GB to 1TB of SAN Storage.  Using a product like Datacore SAN Melody, this is not expensive.  Enterprise deployments requires either more SAN Storage, Snap/Flex Clones, Ardence/Provisioning Server or emBoot's winBoot + sanFly.

There is no free lunch.

Patrick Rouse
Microsoft MVP - Terminal Server
Sr. Sales Engineer, Western USA & Canada
Quest Software, Provision Networks Division
Virtual Client Solutions
(619) 994-5507


This should have read "Customers w/o any Terminal Services or Citrix infrastructure have the biggest licensing challenge."  I'll post the MSFT Licensing Requirements on our blog.



Here is the Link for VECD License info from Microsoft doc. The bottom line is if you want to do VDI you need VECD whether you are running XP or Vista
Licensing Vista with VM

Jim Kenzig

The reason I posted the summary on our blog, is that I've ready those documents many times, but it doesn't give you the information you need to know.  It's very detailed, but at the end of the day, doesn't give me the information I need to take action.  The summary I posted is a one page what you need to license from MSFT to use VDI.


I agree with you. We bought a 4TB SAN the other day for about $7K. It's an iSCSI SAN based on Microsoft Unified Data Storage Server.



The list of unsupported feature is even longer when looking at the bullets on page 10 of the Desktop Delivery Controller Administrator's Guide (CTX116843).

Nevertheless, there is one big plus: Dynamic Client Drive Mapping. It is absolutely awsome to be able to map USB mass storage devices during a session! I am looking forward to see this on Presentation Server and XenApp.

Nicholas (


Well said brian... we need 100 percent to trully kill the traditional deskop. Otherwise im not overly excited on creating another infra to run and maintain for a limited stakeholder count. All i see is complexity, risk, capital build out and maintenance eating into the ROI. :-( 

Im still waiting on a HW vendor to supply hardcore performance data for Vista guests under moderate workloads. There is to much quoting against XP with 256meg of ram. Distributed VDI is looking better everyday.


Hey Brian,
How's the memory lane when it comes to the ICA/RDP "protocol"?
As you have stated in the past there is no real differance between the protocols technically speaking, rather it's about client side/server side features implemented. ICA however have only 32 virtual channels and RDP some 65k.
Is it just for conveniance that you state:

"Citrix's ICA protocol is light years ahead of Microsoft's RDP protocol"

ps. I'm a Citrix guy, so no hasty conclusions of my sayings coming from competing vendors please.

We've been running Server 2003 desktops on Virtuozzo for awhile and were only able to get around 20 "virtual private servers" when running 32 bit Windows. This is a known limitation caused by Windows memory management in 32 bit. Apparently 64 bit is completely different, but we were unable to use it because of application incompatibility and lack of drivers. My main point is that you have to be able to run 64 bit Windows server if you want density.

everything I said in the past was in the context of task workers and business apps. Yes, I still agree 100% that the protocols are very similar, and that it's the client-side / server-side features. But when you're talking about VDI.. when you're talking about a full remote desktop to replace a current local desktop.. that's where RDP starts to fall short.

In other words, in the days of running MS Office and stuff, sure. RDP is fine. But when you try to use RDP to replace your local desktop that does everything, then it's like, "Whoops... this is not the same!"

Cancel bought a paperweight...good luck with that

Nicholas -

Do a search for DynamicUSB utility on Citrix support site:

Allows you to insert USB drives in active Citrix PS sessions by creating a folder in the root of C:\ and populating it with the USB data......nice workaround, free and does the trick...




Brian - I love reading your blogs - youdefinitely know how to create controversy! :) Since you have quoted me acouple of times, I had to post something, and something terse.

We can all debate about whether XenDesktopsupports 99% or 98% or 92% of apps; however, the best test is using ityourself, not worrying about what features are in XenDesktop vs. XenApp. 

I have my virtual desktop that I use, and I worklike a typical office worker. I use standard productivity apps most of the timein my personalized environment streamed into my virtual deskstop, I browse theweb using an installed browser to websites, including brianmadden or Youtube,for good content ;), and finally since I am a marketer I need access to all thecustomer data that I use my LOB apps for (SAP, Hyperion, home grown apps, which are all hosted on XenApp. All I can tell you is that Iam 110% happy with my virtual desktop - it gives me high performance andaccessibility anywhere I want.

So, the real question is not whether the apps are99% or 92% compatible - but how satisfied are the 'OFFICE WORKERS' going to bewhen they get their virtual desktop (in my case it is 110%). And, there is onlyone way to find out, not listening to me or Brian but testing it out yourselffor OFFICE WORKERS:


Btw - when it comes to power users and those who need high graphics performance, Citrix has some plans for those. Check out Project Apollo Tech Preview (that works with XenDesktop) -


Sorry Brian but I must disagree here. If you remember what I said at the MVP Summit in Redmond...

10 years ago, around 1998, we were all probably running Pentium II machines with Windows 98 and Office 97. If you use a stopwatch to time how long such machine took to boot and to launch something like word and you compare that to the latest Intel Core 2 Duo Hyper/Ultra/Super/Beautiful/Pretty laptop available running Windows Vista with Office 2007 and you will notice the boot time/time to load Word is pretty much similar.

What do you learn here? Hardware speed increased so much over the past 10 years and still my laptop performs pretty much like the one I had 10 years ago. The OS/Apps will always use as many resources as possible. Plus with huge (but easy to use) frameworks developers really do not care if their app will use 1GB RAM or 10GB RAM. So in 5 years from now, unless hardware speeds increase way above Moore's law, you will not be able to run more than 40 Windows 7 VMs with Office 14. And I tend to think that will always be the case. Windows 10, to be released in 2020, with Office 20, running on Intel Core 8 Quadruple Octal 300Ghz, will boot as fast as my old Pentium II 233MHz with Windows 98 and Office 97.

Unless, as I said, we have a fundamental shift in technology that breaks Moore's law and gives us a quantum leap in terms of hardware performance. As long as Moore's law is there, Claudio's law (the time to boot and open word is the same regardless in which age/time you live) will be there. And if both laws apply, VDI will never be able to replace TS for two simple reasons: scalability and costs.

Will this kill VDI? Nope. There is a place for a technology that combines the worst a TS has with the worst a desktop has. :-)

Seriously there are many cases where I, myself, recommend VDI. But again, this is to a niche market. I do not see VDI mainstream any time soon, until someone shows me a a server running 150 hosted desktops with real load concurrently.

Cláudio Rodrigues

Microsoft MVP

Windows Server - Terminal Services 

It is humorous to note that the very people that advocate limiting technology are never a subject to their own premise.


When will Provision have x64 support?  How about Windows Server 2008?

Can you create a head-to-head video of Provision's enhanced RDP vs. regular RDP on a typical WAN connection, like Citrix has done for ICA and RDP?  It would be useful to see if there is improved responsiveness (which is what really matters to users, not things like local text echo) from Provision's enhanced RDP.


Hi Sumit,  for this type of workload is it still not more cost effective to deliver a published desktop via XenApp? Higher user density per server and still have silo's for specific apps.

And before you mention the 'pristine' state of your desktop at every boot, I can also do this by using PVS to deliver XenApp!!!!

Don't see the win here under the circumstances you specifically mention.

For a HEAVIER user type then I'd agree that delivering via a VDI solution could be a sensible option.

You are dead wrong Claudio. There will be a HUGE difference in how your apps boot in two years from now compared to past and now. That difference will be flash drives. current harddisks is a legacy technology that's been around for ages and will be replaced by a flash drive which is not mechanical. This in itself will improve things like boot times and app starts dramatically.
Summit, I hear you out but I really don't trust you as you work for Citrix so obviously your response has to be viewed through that prism. Perhaps you are honest and what you are saying is true but one would be mistaken to think you're neutral in your views and opinions.
I like the idea of centrally managed pristine desktop(s) combined with appropriate GPOs and integration with XenApp. 

However, I'm not sure why I would go with XenDesktop over Provisioning Server for Desktops.  I don't see much value there.  In some cases, I would have two ICA connections, one connecting to the PortICA and the second to my XenApp servers.  And as Brian noted, PortICA has some serious limitations right now. 

What are the limitations of Citrix Provisioning Server for Desktops?

Citrix Provisioning Server for Desktops can be a good solution IF

1. You have relatively homogenous PCs hardware; you can have some heterogeneity, but too much means headaches uses the Common Image technology that allows you to support multiple driver sets in a single image.  With XenDesktop, ICA obviously allows you to support just about any end-user device you want.

2. You have good bandwidth (100 Mbps minimum) between the Provisioning Server and PC's  

Provisioning Server for Desktops works well if these conditions are met.  Wyse uses this technology in its Streaming Manager solution, and Dell uses it in their Flexible Computing offering.

A good solution is actually to combine both solutions...i.e. Provisioning Server for Desktops for LAN users and XenDesktop for WAN users (or those with unmanaged devices).  The same image can be maintained for both since XenDesktop uses Provisioning Server under the covers.   

 totally unusable for the general end user population.

Just a short comment: There are significant differences between RDP and ICA, such as how per-channel compression and encryption is handled. But there's no major difference between the number of virtual channels. According to the official RDP specification published by Microsoft "A maximum of 30 virtual channels can be created at connection time" (fo rmore details search for MS-RDPBCGR.pdf). ICA supports up to 32 virtual channels per connection.

There can be up to 64k connections between a TS and clients, which leaves enough headroom for real big TS server platforms. But this number has nothing to do with the number of virtual channels per connection.

Since Windows Vista/Windows Server 2008 RDP 6.x has one advantage when compared to ICA: it supports the concept of dynamic virtual channels. This allows the creation of new virtual channel connections during runtime without pre-creating them during the connection process.


Everyone has an agenda - even Brian.

Sumit - I am considering a move to XenDesktop, what do you use to access your hosted desktop? Don't need brand or anything, just want to understand the form factor - is it some sort of thin device/laptop/desktop system, etc.?



Well said, Claudio.  I've been saying the same thing for a long
time.  The only way you ever see a sustained performance benefit is if
you use your new hardware to run only your old apps, which of course no
one does.  The latest and greatest hardware at any given time will
perform well with the current software of that time--even stuff that's
newly released.  It's what happens in a year from that point when the
hardware isn't the latest and greatest but the software is even more
resource-intensive and bloated.

I agree that flash drive technology will alleviate some of this problem, but I have a hard time believing that current hard drive technology is the only limitation.  And even if that were the only limitation, then some other limitation would take its place, such as cpu or bus speeds.



Of course nothing is 100%.  Brian, I'm a little disappointed that
you even suggest that a less-than-100% applicability rate is a
shortcoming.  Since when has a new computing technology ever had to
achieve 100% applicability to be viable?!?  Off the top of my head, the
only thing that comes close is Windows, but even it tops out at about
90% of the client desktop market.

As Mark stressed during the
Synergy keynotes and as Sumit evidences below, VDI (with XenDesktop at
least) provides typical knowledge workers with solid-performing,
customizable desktops that they can access from anywhere regardless of
their client device.  Looking at it from a microcosmic view, even if
VDI could run only 90% of the apps Sumit needs, that leaves only 10%
that need another delivery solutions such as XenApp/TS.  With the rest
of the 90%, you've now centralized management of both the app and the
desktop, while providing the user with a customizable and
device-independent solution that can be accessed at the office, at home
or on the road.  You might ask, why not remote all the apps?  Well, now
you've relinquished centralized management of the desktop.  To that you
might reply, then why not a TS desktop?  Well, now you've taken away
some customizability and impacted the user experience.

But of
course no solution will work for 100% of apps or 100% of users--just
ask any admin who has 3ds Max or Lightwave users if he or she expects
VDI to impact those users (though Citrix definitely is making
impressive strides with graphic-intensive apps!)


Provisioning Server is awesome technology--who says you have to go with XenDesktop over it?  Heck, why even choose between the two when you can use both together?  Have your VMs boot off a Provisioning Server vdisk image and use XenDesktop to deliver the VMs to users.  That pretty much eliminates the hardware homogeneity and bandwidth issues, no?
.. at the cost of datacenter build out. Targeting (as other guest said) WAN or unmanged machines reduces user count. Its all about trade offs... also provision to lan user allows 3d until more protocols suppport it.
Again I disagree. Have you ever deployed Citrix or any other technology with Flash Drives? I have been actually working with these for YEARS and they will NOT change that much in the next 2 years GUARANTEED. One of the major Citrix farms I have worked on had flash drives on ALL the servers for paging (agreed it was not for everything but for paging already helps a lot). Did it help? Oh yes, a lot. Was this an improvement that would change computing and computer power as we know as of today? NOPE.
Just try a MacBook Air with a flash drive or any other machine with that (what I can tell you you have not tried). This will not change anything. And if it does, developers will find a way to use all that new boost in speed to create even bigger and fatter applications that will perform as fast as my old Office 97 on my PII box. That is the part you do not get it.
Developers will ALWAYS find a way to maximize hardware utilization what will keep the perceived performance at the same level as it is today.
I will say it again: unless we find a way to fundamentally change what we know as computers today, Moore's law and Claudio's law will still apply. WIth or without flash drives.

Cláudio Rodrigues

Microsoft MVP

Windows Server - Terminal Services


Brian, I had the pleasure of seeing you sing and dance at the Pubforum in Amsterdam Brian, and what i took home with me there from your session was the thought of changing the way we work, not focusing on XenApp version X, Terminal Server version X or desktop deployment solution X. But the notion of having more and more ways to reach more users and more scenarios made me think of the whole delivery idea in new light.

To me this whole debate is kind of crazy, who cares really how many percent of the apps out there we can provide on a XenDesktop, or Provision Networks Desktop? or VDM? To me XenDesktop represents new possibilities for my customers to deliver more services to their users on a known platform that they trust. I think most of my customers would like to have their users running 100% TS/XA but that is not an option.

VDI however opens the door for them to bring more of the users not able to live on TS/XA into their datacenter and some of them onto their virtual infrastructure thereby supporting the initiatives they have done to save cost, manpower and rackspace.

And there is another trend i tend to see when talking to customers about VDI. They see a new chance of emptying their offices of desktops, thereby giving their employees a better work environment, less noise, less heat(cooling needed), and more Desk-space. These are simple joys but to the office worker who just wants to work in a simple way this matters. And when hiring new employees, these things matter.

I know you can use thin clients on traditional solutions but what i am talking about here is getting some of the workers that we did not get 100% on traditional platforms to now be able to join the new world and enjoy the benefits.

just my two cents.


I know, I'm just saying it's possible.  You'll have data center build-out with VDI whether you your VMs boot off Provisioning Server images or not.

not if you use distributed VDI :-)

Guess you never had a Windows XP embedded client that run on flash drives have ya dude? We have and it really isn't faster. Vista precaches apps in memory and guess what they still don't load any faster either. Get a clue.
I would suggest that VDI adoption rates, even if only a few percentage points over the next few years represent many millions of desktops and new opportunites for all to expand. This means investement and innovation will continue here. VDI is unlikley to be cheaper than SBC for some time, (storage costs, OS costs, etc) and VDI protocol limitations will not address enough use cases to replace all physical desktops. So Niche use cases, perhaps, but VDI will address many new use cases that the other solutions don't offer. Power Cooling managment, centralized mgmt with personal control i.e. developers can reboot blah blah we can all figure this stuff out. As for the protocols, RDP for just an App like MS Office vs Desktop is a bit of a stretch. Just try using Outlook and working all evening over native RDP over anything but a high speed LAN and see for yourself. ICA makes a difference and Provision add some cool stuff. SPICE, NET2Display, Calista etc are innovations in progress so we'll see. Also all the solutions are trying to sell at least from my view, their integrated everything approach to this problem. Does one really want to be locked into vendor solution X. Just look at the VMWare SUN VDI thing today. Does one really want to be locked into deploying SUN hardware to deploy VMWARE VDI to address their protocol problem with RDP? If they smartened up (especially the smaller guys including Citrix)and better enabled their products to interoperate with other solutions, we could choose the various layers of a solution and put them together as one see's fit. That goes up the stack, hardware, protocol, mgmt tools etc. This would make the adoption of VDI less disruptive and presents new models for the various vendors to potentially explore new addressable markets for a broader use case than they are considering today. Again I assert only a few percent of the worlds desktops will use VDI over the new few years, so plenty of room for those with vision and courage to open their eyes and figure out ways to capture the larger pie. Great blog, Brian, Sumit both make valid points and great to see so much dialogue on an emerging model that many of us will benefit from.
What is your reason for running "Windows Server 2003 desktops" using Virtuozzo instead of using Windows Terminal Services?

Caludio, you did buy the LinRDP and dosRDP from some sobscure chinese firm that based their products on the GPL rdesktop without ever releasig the code yourself nor your chinese vendors. It's in the open. What was your thinking?

ps. Not from cendio nor any other from the rdesktop lists.

Sorry - could not reply back earlier - traveling. 

Why use XenDesktop for office worker workload instead of XenApp? 

Can we twist around XenApp to deliver rich personalized desktops. Yes - we can. However, the platform is not optimized for that environment - it is best suited for running locked down apps. Even the desktops that get published via XenApp get locked down as much as possible where zero personalization is allowed.

XenDesktop makes personalized workspace for virtual desktops a bit simpler. Office workers get not just a familiar user experience as a physical PC but can access it from anywhere.

You know what's interesting to note here - clearly there are several camps on whether desktop virtualization fits for the use case or not. This illustrates the problem - the problem that we all have different types of users that we 'service'. Between XenDesktop and XenApp, you now have an opportunity to serve more users centrally. There is no SINGLE answer for everyone - you will have to think about your user requirements, cost of personalized workspace and then make a determination who is your 'TASK WORKER - SERVER BASED DESKTOP' and who is your 'OFFICE WORKER - VDI BASED DESKTOP'.

We can help - I will write up a blog with details on how to profile different users after I return from this LONG trip.