Citrix, Microsoft, VMware: The GOOD and BAD with each of them in EUC right now

There's a lot going on in the EUC space right now! One of the most-asked things I got at VMworld last week was some version of "What do you think of [Citrix | Microsoft | VMware] right now?

There’s a lot going on in the EUC space right now! One of the most-asked things I got at VMworld last week was some version of “What do you think of [Citrix | Microsoft | VMware] right now?” So I figured I’d make a quick bullet list for each company of what they’ve got going on that’s good and bad.

Note that this is only in the context of EUC and desktop virtualization.


The Good

  • Biggest market share
  • Huge install base
  • Most mature products
  • Still very much synonymous with “desktop virtualization"
  • Best / most features
  • They say “on premises"
  • Best protocol (yeah, I said it)
  • They’re not trying to make their own cloud / not competing with partners
  • Citrix Workspace Cloud
  • Citrix Workspace Cloud (seriously, nice work there)

The Bad

  • Elliott
  • No CEO
  • Talent flight
  • Partners are pissed
  • XenApp 7 debacle
  • Product quality issues
  • Octoblu (cool, but too much focus?)
  • Distracted
  • Competing with Microsoft on too many things (XenServer/Hyper-V, XenMobile/InTune, ShareFile/OneDrive)
  • First class integration only with other Citrix (non-leading) products
  • Lack of momentum. Tired.
  • AppDisk (seriously, just buy FSLogix already)


The Good

  • They own Windows
  • They own Windows
  • Office 365
  • Azure
  • SCCM (They did a good job convincing the world they need this. Huge install base now.)
  • Windows 10 (Everyone will go to it, easy to manage, it’s a joy to use)

The Bad

  • Most people hate them.
  • Ok, “most” is probably too strong.
  • Ok ok, “hate” is also probably too strong.
  • No desktop SPLA
  • Still screwing customers with their monopolistic actions
  • People buy them because they have to, and that’s never a good thing


The Good

  • They’re saying all the right things
  • Momentum, buzz, etc.
  • Project A² (They can manage local apps on physical desktops)
  • AirWatch
  • Shawn Bass
  • They own the infrastructure & platform in many (most?) companies
  • Synonymous with “Virtualization” and “VDI”
  • Everything they’re doing with micro VPNs and NSX is just awesome
  • Windows 10 hug from Microsoft

The Bad

  • Flex?
  • Enzo?
  • Horizon 6 still has a long way to go
  • They’re owned by an old school storage company
  • On premise? Seriously?
  • vCloud Air? They’re competing with AWS and all these other massive cloud providers? They’re competing with their own customers??
What am I missing? Post to the comments and I’ll update the list...


Join the conversation


Send me notifications when other members comment.

Please create a username to comment.

Nice one Brian, straight to the point ;-)


one point...

Why do you have Octoblu as a "bad" - I think it is a huge positive


Regarding Octoblu listed in the "Bad" column for Citrix:

First, I think Octoblu is cool, and I get that Citrix needs to make R&D investments that seem illogical to outsiders like me but that will potentially pay off in the long run, and I understand that the R&D net is cast wide since you don't know what will pay off and what won't.

I listened to the earnings call when they bought them. It wasn't a lot of money, and I heard Mark talk about how it's about integrating "everything" (IoE) and it will allow partners and customers to extend the smarter workplace, etc.

So I get all that. And I agree with that? It's cheap. It's cool. Why not?

The reason I listed it as a negative is that I felt like at Synergy, there was so much talk from Citrix folks about Octoblu and not enough talk about some of the reason issues. Like a magician's misdirection. "Pay no attention to the real problems we're not talking about... Hey, did you see we can automatically call people who aren't in meetings???"


ok. I generally try to not jump in here, but to say AppDisk - Buy FSlogix (dont get me wrong there may be a reason to buy that) shows a misunderstanding of both products.

Yes both have a file system filter. But so does anti-virus apps, and View persona... neither of which do app layering. Kevin has some cool stuff going on in the hiding of apps and what not, but somehow you are thinking it is app layering (and all the accompanying stuff like registry merge, conflict resolution, etc, etc)

VMware has a big vision. They are telling a great story. But bundling up existing tech like Mirage and VMW Player and calling it a solution to something is really just check box filling.

Citrix of course has its own struggles as of late. There are many reasons for that (some listed).

best comment in this article is people buy citrix because they have to...  VMware has believers.  Much like Citrix used to have. And I dont mean believe they have a better product. I mean believes in the vision, the story, the ideas.  IF VMW can execute its a hell of a story.

But then again, we have to see analysis of tech that actually talks about what it does or doesnt do and why its important. Most analysis we are seeing these days only relies on the power points and talking points. not the nuts and bolts of all the bullets in the announcements.


@Ron, thanks for jumping in!

I just scheduled a call with Ron to talk details. Look for an article next week "Brian's misconceptions about layering." :)


I like the VMW marketing pitches, but at the end of the day, Citrix actually does it.  I do agree with Brian when he said the Citrix keynote wasn't what I would expect from a company who is going to lead me into the sunset (I'm paraphrasing, I forgot his exact words), but it wasn't professional at all.  It could have been a lot better and it should have been with their track record of the past 18 months.  CWC is cool, pricing though, not so cool.  I run XD on premises (as Brian puts it) and I already have a sunk cost, now to move that to CWC is going to cost me $20/u/m or $240/u/y, that's alot of money when you are talking 100 users or more.  Sure it's cool that I can then burst to AWS or Azure, but then I gotta pay them as well on top of that CWC management.  I could just go to Desktone / Horizon Air for $35 or so per month and be done with it and come out less.  Sure it depends on use cases I get it and everyone is pushing cloud as there is customer lock in, no piracy, etc, etc.  M$ is going exchange in the cloud on their next version as the only option and it'll probably bump up revenue b/c alot of people probably aren't honest on their mailbox licensing, but with O365 you have no choice but to pay the piper.  Adding up AWS and CWC will cost some $$$ at the end of the day.  Cloud has to be economical at the end of the day, I should be able to get value and features.  If I'm already paying Citrix for SnS who cares about the upgrade process, I can just call them and get it done if I get in a jam.  Would it be nice to only update the VDA, sure it would, but at a cost of $24K for 100 years per year, maybe not.  $24K will buy a couple of Supermicro 2U beefy hosts to run tons of stuff on.  As for Appdisk, I'm in the camp that Citrix should just buy someone, LWL, FSLogix, etc, etc.  However combining appdisk with appdna would be pretty cool if it is supported.  I sure hope that appdisk is available to all XD editions just as PvD is.  AppVolumes is really cool in how it can attach the vmdk at login for non-persistent disks.  Appdisk at least need to have the same functionality to be competitive, perhaps that is why it hasn't been released yet, hopefully the finishing touches will be put on it soon.  Time will tell here if Citrix can woo us back.  


SCCM is good for Microsoft but is it good for all those companies invested in it?  Seems to me like App deployment and management is changing and the options available growing....anyone brave enough to ditch SCCM and go in another direction?


Agreed with the fslogix comment. Would be a great addition to the Citrix line up.

Interested in seeing the article next week Brian as I thought with fslogix 2.0 they now can do "app layering" aka "app containers" and "profile containers" at the windows level versus the hypervisor in addition to the core of the product which is the system filter.


We should not confuse App deployment with Application layering or application packaging. SCCM is a delivery mechanism. If I am not mistaken AppVolumes works with SCCM in the same way AppV can.for example.

Anyone can create a layer in Windows - simply create and mount a VHD using disk management and you can save your profile or apps to that location,. The money, skill and value are in managing those layers and working out how the hell you can ship them around the network...

I would also add there is NOT one solution that fits all. Windows Store apps and Business store apps also will be part of a desktop at some stage and placing all your apps in the gold build, in a layer, or all into AppV is just not possible. Application delivery will be a hybrid of technologies.


"Anyone can create a layer in Windows - simply create and mount a VHD using disk management and you can save your profile or apps to that location,."



Have to agree with Ron here, think there is far more to layering than just mounting a disk and presenting different files to different users.  

registry is pretty major on anything, and issues with conflicts between layers, prioritising them etc.

but you're right, anyone can mount a vhd and do that.  becomes more complex once merging is required.

also look forward to Ron & Brian's post.  just wish unidesk would support server os/xenapp (dedicated and shared) type environments.



I'm going to brave enough to ditch SCCM. I have it down no our internal road to have SCCM replaced with an App Layering solution. But in reality it really boils down to whether or not we can virtualise all our user base. If I am keeping some physical desktops out there - I still need to manage them in some way.

I have had a quick look at UniDesk and FSlogix, and I agree they are 2 different solutions. For me I am looking to simplify our environment and operations. Make things as easy as possible! We are constantly being asked to do more with less resources. So many people I know are saying we do not have enough people. So let's simplify things. Why do I want members of my team spending hours on trying to sequence an application. Even with FSlogix, you are looking at hiding the app. What about licensed apps (Adobe's, MS, etc...) Even though it's installed and you are hiding the app will they officially recognise the use of FSlogix as a valid mechanism to controlling what apps users get and what licenses they consume?

With FSlogix it still seems to have to understand the app - file locations/Registry paths etc... why would I want my guys to waste their time effort in doing this - it's just a ball-ache I don't need.

Citrix should buy UniDesk!!


@Simon2000 I believe you highlight my point that there is "no one solution" that fits all when it comes to deploying OS, Apps, Data and users. In addition we cannot ignore non virtualised platforms, aka traditional desktop, laptops and new tablet devices…EUC should include them all IMO.

To simplify and reduce the operational cost of EUC requires solutions which can help manage OS builds, Applications, Data and Users across the entire estate, regardless of how they are being delivered.

Today, IMO and unfortunately, no one solution or vendor is able to manage every aspect of a users desktop nor work across the entire estate of both physical, virtual and cloud. Offline machines, remote locations, hub sites, DR scenarios, Windows 10 cadence, compliance, security threats and application deployment continue to challenge every technology in one way or another.  

Some of the newer technologies show great promise and are starting to prove that new ways of managing an aspect of EUC are possible. AppVolumes (and hopefully AppDisk) are good examples of this when it comes to deploying applications – it’s a new approach compared to packaging and AppV and they can reduce the cost we associate with sequencing for instance. However, many of these technologies were designed to solve challenges at just one level/layer (eg. apps in this case) and more importantly were born to solve different challenges or solve use cases on just one type of environment (aka VDI/TS). AppV was the same, (isolation not deployment), and SMS (for those that remember? :) the same. They were solving different problems and, or in the case of SMS, were before we had our desktops virtualised

To date I have seen Unidesk be successful in delivering and managing the OS and applications layers across both physical and virtual, and AppSense (for whom I work for) has some great joint case studies where we have worked to help people achieve “single pane of glass” management of OS, Apps, Data and User across a heterogeneous estate. This provides great benefit in what I call the “Hybrid Windows world” where users are jumping from one platform, OS or app delivery method more frequently than ever before. It reduces management costs yet can provide a seamless user experience.

The ability to have a single pane of glass which can dynamically deploy, configure, secure, optimise and audit every layer and across all platforms continues to be a challenge for many and today, I believe still requires more than one solution.  

The cost of managing multiple solutions can be an issue, but again we should ask each technology how it integrates with other systems and how much can be driven by other products or processes. For example, some of our customers  script our agents/config and utilise our API’s + Powershell cmdlets so that this cost is reduced and management can be integrated and automated.

If I may I would also suggest that you also consider User environment management solutions like RES, LWL and AppSense, as these provide far more than just user profile management in a Virtualised environment. They provide important functionality like policy, application configuration, application whitelisting, security, user rights, privilege management and auditng which compliment how the OS and Apps are delivered. These technologies also support tradition desktops and in many cases the number of OS images and number of application packages are significantly reduced when you remove the customization, policy, security and personalization away from the underlying OS and apps. This, to your point can help reduce cost associated with application packaging and OS image management.

The good news is that we have choice and in most cases the technology all sits nicely together… it may not be as simple to deliver as we would like but then we are not just buying desktops and deploying a few hundred locally installed applications on Windows XP for all our users anymore?



You didnt know we support Server OS's? XenApp and straight RDSH 2012R2 (and can support 2008 R2 if there is interest).

Happens a lot with us. people look at us and dont see something then never come back to check. We dont have the Citrix or VMW megaphone to put new features/supported configs... Then again that megaphone often comes with promising the moon. Maybe I dont want it.


Geez I guess it's ok to hijack the comments section for infomercials on now changing the topic of a post. But hey free speech and all that.


@Ron, every time i've spoken to unidesk it has been "we only do vdi with workstation OSes".  though to be honest i haven't asked for a year or so.



yup, introduced March/April this year.


Why do we still talk about layering only for VDI (or even Terminal Servers) and not for physical desktops? The same management benefits should apply there. Right?

The old reasons for storage savings etc. are less relevant today with all kinds of de-dup built into storage.


@Simon Townsend - Agree about real UEM needed from AppSense or Liquidware Labs.

@Ron Oglesby - agree with the HAHAHAH, it also irks me to hear, "anyone with a file system filter driver can layer apps, that's the easy part."  It's not.

@Amitabh Sinha - Spot on about Physical, why should it matter - Not all the layering solutions support fact only a couple. We're introducing support for physical desktops with FlexApp now, big advantage.  And physical is a gateway drug for VDI :-)

@All - if Citrix PVS/MCS is important to you, make sure your layering is compatible.


Love this article. On the Citrix side, I think you forgot one thing: the lack of a real UEM (I know UPM is up to version 5.2 but its still ancient!) .... I think they should have bought liquidate labs a long time ago, along with FSLogix  ... but with Elliot in the mix trying to sell assets, I don't think Citrix is in a position to buy jack sh$$t. And why on earth would you buy octublue when you're in that much trouble?