Microsoft adds App-V for TS to the RDS CAL!

See this blog post from Microsoft about the changes in TS CAL licensing. In particular, they will now include the App-V for TS into the new Windows Server 2008 R2 RDS CAL!

See this blog post from Microsoft about the changes in TS CAL licensing.

In particular, they will now include the App-V for TS into the new Windows Server 2008 R2 RDS CAL!  This doesn't get you an App-V for windows desktop CAL, which you would need for VDI.   Price of the CAL will go up at the end of the year by 5%, but is discounted at old price until then. 

More info in this blog:

And in this:

Join the conversation


Send me notifications when other members comment.

Please create a username to comment.

Curtains for Citrix streaming I guess for sure, unless they are going to do something different with what they have. Hurts ThinApp in a similar fashion, Appstream + SVS, no need for the price. I get it, brilliant move by MS, but at the same time I hate it since they are trying to draw the world to do only App-V. They will kill competition in this area now, and move even slower. Somebody has to do something different. Go XenoCode!!!! Show the world the value to agentless done right without a client from a browser.

Sorry Ruben S and co, who keep saying that agentless is useless without management. You forget that agentless integrates with other systems. SO PLEASE get the message out as well, that this is all designed to marry you to SCCM. WAKEUP, MS are a monopoly, we will get screwed unless there are real options.

What's telling is that they could not give it away as part of MDOP, which clearly also tells me that RDS is not part of the Client cool ground and this is bad for VDI. TS should be part of the MS Client group period, not some outcast server team. Wake up Mr. C of Terminal SErvice at MS or can his boss please wake up...

I guess in Windows 8, this will become part of the OS, and they figured they could not sell MDOP then. That's good news, because then they can fire that idiot women who did nothing with Kidaro and got promoted and runs MDOP. There is no other value in MDOP, and once App-V is a commodity in the OS, what the heck are they going to sell. Hmmm I see another shopping spree, anybody want to do a start up and sell to idiot women in charge at MDOP!!!!


You have to specify Ruben S, as if there's any other Ruben that matters?  ;)


There are from the same place :-)


Ruben S. has way better hair (says the short bald kid)

BTW, is DirectAccess really just a DNS based vpn by another name? Overhead? Does it matter?


How awesome is it that the new RDS CAL is *technically* a 2008 R2 CAL, and 2008 R2 is x64 only which App-V doesn't support yet? :)

I know, I know... you can use these for 2008 "R1" access, and any existing 2008 TS CALs now instantly have App-V access.. so this is really a good thing.

But still, it's funny!


At least the 64bit client is coming with 4.6 finally.  (You may remember that we showed it as an exclusive at Briforum).  Unfortunately this is still listed as a 1H2010 release, at least at this point.

To be clear, this change does not affect App-V for the desktop, which is still part of MDOP.  Application Virtualization remains necessary for VDI and leaves an opportunity for all the players.  Of course since you already will need VECD (or whatever it gets renamed to), getting MDOP doesn't seem all that expensive.

If you use a different solution for VDI (and assuming all those companies actually go to vdi), then staying with a single application virtualization vendor will probably be more important than client licenses for the TS servers.

But thinking about the service provider model, I don't know if ASPs get this same deal, but if so those folks should be chearing big time.


Whoa appdetective sure does have it out for that PM @ MS MDOP LOL.

@Trex - Direct Access is an IPSec VPN with fallback to IP/SSL VPN tunnel.  What was that question in reference to though?

@articlecontent - I'm excited that MS is bundling the App-V CAL into TS CALs.  I'm less excited that a customer of mine already bought App-V TS CALs.  Though they shouldn't feel too bad since they're using 2003 TS CALs and this CAL change pretty clearly spells out 2008 TS CALs which means more $$$ anyway, so oh well.

@myself - Stop being schizo.



With regards to appdetective's comment...forgive my ignorance but hasn't MS put out the App-v lightweight streaming server so that you DON'T have to use MS stuff in order to manage it (Powerfuse/Appsense) for example?


@A.Nikolic - Yes LWS does exist that allows you to not need management infrastructure.  I think what appdetective is asking for is a clientless virtualization layer.  Xenocode and ThinApp both have this capability.  App-V, SVS, Citrix Streaming all require a client installation in order to use the virtual apps.  I can see pro's and con's of both approach, but I don't see MS getting there any time soon with their kernel components in App-V.  They'd pretty much have a rewrite on their hands.  Unless of course they put out an App-V kernel component into Windows Update (muahahahahahah - seriously I doubt that'd ever happen)



@ Shawn..

I agree 100% about the pros and cons of client vs clientless virt. layers but I fail to see how, as Appdetective stated, the Strategy of MS is driving us towards SCCM when there are other options available for managing your App-V environment .

<fanboy mode on>

BTW big fan on the stuff you have done on Briforum and your blog!

<fanboy mode off>



App-V suffers from a few things. Firstly it's a single service to run all your apps. I hate that as much as I hate brokers as single points of failure. The dynamic context stuff is too complex. The agentless guys InstallFree, Xencode resolve this by dynamically loading all the stuff at run time from seperate packages. Think smart shortcut. So you don't need one big package, which is a lie spread by MS. I believe Citrix Streaming can do this as well, but need to verify. App-v streaming is very average over WANs in my experience, although to be fair I need to do more testing, but I can tell you for sure that xenocode streaming is very good based on extensive testing.

Microsoft will keep adding features to SCCM and taking them out of App-V, inventory, reporting, application portals etc as the world moves on to make sure that App-V works best with SCCM. So if you are not a SCCM lover your should think twice about App-V.

As Shawn points out, App-v is fundamentally flawed in that it's at the kernel level. So guess what you need admin rights to install it. What I want is applications to be much easier to delpoy which is why I like the Xenocode, works on most Windows OSs, fast streaming, agentless, light user mode agent for browser plug-in and so simple to use their packaging templates and a very low overhead on their virtualiztaion layer and runs almost like native. In other words rock solid implementation, because they have focused on the core, and not diversified into management that takes resources and focus away from bulding something rock solid.

I can take Xenocode and embed it into anything I like. So much more could happen with their security wrapper etc to drive consumption of apps in so many new ways. Microsoft will not build that piece unless it's ties to SCCM/Azure or similar and lock out others from doing anything once they are married to App-V. I want FREEDOM to do what ever the F I want to do with MY apps and not be locked in by MDOP idiots etc  who have no clue and innovate at the pace of snails.

We are all killing choice in the application virtualization market by following MS blindly. There is nothing exciting about this for me, just fear that once they control application virtualization there is no incentive to extend and they will force this arhitecture to upsell ONLY ms products. That's what they do, just look at their VDI licensing and they are slowing that market down as well. SO BEWARE.


@ Tim

Yes any Microsoft Hosting Partner leveraging the Service Providers Licensing Agreement (SPLA) gets the same new rights too!


@ Alex

Thanks for the clarification for SPLA subscribers! The CAL for App-V on TS under the SPLA program was dirt cheap anyways...



Thanks for your extensive answer..I see your point now..thanks!


So with the improvements to application publishing with RDS and free App-V TS Cal's what left for Citrix value add?



We need more noise so that the product group understands SCCM isn't everything to everyone. SCCM isn't a real-time system and never was. Until MS does some serious rewriting of the SCCM architecture it never will be.

Aside from that there are some people that just don't want SCCM in their environment. (i.e. TS Admins)


@Aaron - App-v  for servers only manages the application packages on the TS Servers - so only competes/replaces streamed apps to the TS server. The Citrix stack still provides the high performance protocol to deliver the app to the user, the broker, the scalability etc.

This isn't a concern for Citrix, and I personally believe is a great move as it provides choice for TS / XenApp administrators as to how they manage streamed/isolated apps on their servers. Of course this doesn't have any effect on installed apps on TS, which is how most TS apps are provided.


@Kevin Kaminksi,

I agree..I'm currently converting our TS farm with Citrix, RES Powerfuse and Softgrid 3.2 to use the Lightweight streaming server 4.5. We are  pretty happy with how we can leverage Powerfuse to use the Sofgrid apps instead of adding another layer (eg SCCM) to manage this.

Every software layer you add, is another layer of bugs added, this is one of my mantra's :)



I don't think MS plans on marrying SCCM with App-V so that you couldn't run one without the other anymore. Built-in reporting has lost some of the reports, but let's face it: the reporting in App-V has always been little bit lacking so for serious reporting you would need other means anyways.

Besides, when it comes to webportals or these other things, 3rd party could create solutions for those (like Login has) and then you would not need to use SCCM (or System Center family). I have actually always been puzzled why there hasn't really been such 3rd party add-ons more..


I think this will mostly hurt Citrix. Next step i guess from MS will be adding App-V license to client CAL (like  SCCM client).

Like it or not this is the pattern MS has been using. I see MDOP like Vista Ultimate Extras, it will go away (vanish). It is here to just sell SA.


@Kalle (although he really knows this)

The lack of 3rd party add-ons for App-V in my mind has been the lack of commitment by Softricity (and so far pretty much carried on by Microsoft) to provide well documented and stable interface points.  

We built it to have good interface points, but in the rush to get the product built those interfaces were not documented.

Much of the first two years after I left the company I pushed them hard to finish the job on the interfaces.  They never came.

In fareness to MS, they did finally document the SFT format (thanks to Kalle) and the command line sequencer.  But they have not done so for the PKG.  Also, they never got past a draft on the plug-in dll interface to the provider pipeline (which changed before I even saw the draft), or document the "management interface".

A robust third party environment requires well documented interfaces and here Microsoft has pretty much failed.


To add to Tim's point he's right and this just underlines my fear. MS app-v innovate too slowly and are simply trying to get App-V out there to control this market and us blindly following their option and not creating choice is going to hurt us. They want to build their own stuff and not enable 3rd parties. By controlling the application virtualization engine as a monopoly does and not opening up interfaces fast enough, they control the pace of the market allowing them to build their own good enough solution in their own time. App-V is a flawed architecture in many ways, and there needs to be an alternative that is open allowing people to extend and solves for many of the use cases that MS doesn't.

More power to the InstallFree, Xenocode, Appzero  and even ThinApp, innovate, innovate, innovate. There is a market, and we need new ideas that are better than MS, and move faster to enable application virtualization solutions not flawed by single point of failure architecture.



While I understand your points, I will never agree that an installed client architecture is flawed.  I believe that both client and clientless flavors of application virtualization have different bennefits.  In particular, I find the "fear of a bug in the client" does not match the experience of eight years of history of the product.  And since clientless does have a client that is run with each application, a bug there affects each of those apps too.

Alternatives certainly do help improve products.  I like what some of the guys you mention are doing (although Appzero is quite a different animial) and hope that all the vendors can improve.


@Tim, I'm arguing that the installed client architecture represents a single point of failure. For example, the service dies and all the running apps that depend on that agent die. That's where I take issue. It's like saying every time IE crashes my office apps should also crash by design. It makes no sense and the fact that MS still hasn't fixed this shows they just don't it.

On the pure client side the fact that you need admin rights to install the damn thing, just limits it. We too easily just accept crap like that from Microsoft.



Well we just aren't going to agree on that!  Single point of failure is a great paper argument, but then LSAS is a service that represents a single point of failure in the OS itself.  Don't think I have ever heard anyone complain about that one!

As to needing admin rights to install, I suspect there are many IT folks out there that view needing admin rights to install software as a feature!  It's all about the problem you are trying to solve.  If you have a different problem, then a different tool might be the answer.



Nope we won't agree. Bad architecture is what it is. Forcing independent application threads to now depend on a management thread to execute, changes the reliability model period. It's crap. Apps have always run independently and Softgrid/App-V breaks this model with bad architecture. Comparing this to LSAS is not apples to apples as that's always been the case.

As for the admin thing, you don't have to apply permission on the end point and screw up reliability. With agentless you can simply run from a network share with permissions applied. Anybody could be enabled to do that with automation, and we can reduce packaging and management overhead on centralized IT teams.There's many ways to skin a cat with single image apps....and how good is the dynamic suiting in App-V anyway. Another dumb way to implement.

I'm ok for anybody to disagree, I'm just think it's important for people not to act like sheep and follow MS without understanding what they are loosing. Tim not personal, I get your love for App-V given your history, that's great. It's MS  that is screwing the industry.



You talk about innovations by the other app virtualization vendors, but the truth is that when technology is bought by the big company (like has happened to App-V) then development pace tends to slow down significantly. You cannot disargree that this hasn't happened to ThinApp as well!

Also, when it comes to application virtualization as a concept, there's only that much you need to do before it's getting more hard to add anything of real value. Those products that seem to be innovative and fast moving are actually more or less catching up with the more established players, it's pretty given that they need to roll out new versions very often.

Well, each to their own. We all look things from our own angle and there's no one universal truth when it comes to client vs. clientless etc.