[Note from Brian: We originally published this article the day before VMworld last month, but I think everyone was focused on VMworld and so no one read it. :) It's an interesting idea and I'm curious about peoples' comments, so I've reposted it to the top of the list for today.]
Well it’s VMworld this week, and I’m looking forward to learning about what’s new in the EUC (End User Computing) business so we can hopefully start to move beyond having RDSH vs. Non persistent VDI debates and even lighten up on RDSH.
Brian’s recent RDSH versus non-persistent VDI blog left me wondering: do we spend way too much time incorrectly framing a discussion and not empathizing enough with customers? From personal experience and talking to many people, I think end user computing is a multi-dimension puzzle that many people don't appreciate. So I thought a framework would help highlight some of the conundrums customers face.
I am no industry analyst (and I don't pretend to be one), and I don't have the resources to do research across the whole industry. Regardless, I was curious enough to come up with my own version of what I call the "customer-focused end user magic quadrant." Of course I was inspired by the Gartner Magic Quadrant, but my customer focused end user magic quadrant is based upon my own experience and interactions as a customer, selling to customers, and consulting for customers. This is not the same as the mighty Gartner Magic Quadrant, so my Bush League one requires some explanation. Hopefully the Gartner lawyers will be just dandy with that! :)
Before I begin to explain my customer-focused end user magic quadrant, I should highlight what it is not. It’s not meant to be a way to assess vendor specifics against each other, although you could use it to form a high level opinion of offerings. Instead I put together a way for enterprise customers to map end user architectures and solutions to their needs. Within each of these architectures/ and solutions you'll have to decide what to pick (unless I am naming a specific solution in a quadrant). I’ve always liked the idea of crowdsourcing, so I’d be delighted to receive feedback to refine the framework.
The quadrant illustrated below has four axes that represent what the enterprise customer wants. Then various end user computing architectures/solutions are evaluated against these axes. Below is an explanation of the axes, followed by a discussion of my thinking on how I placed various architectures/solutions into their respective quadrants.
The Customer Focused End User Magic Quadrant
Bottom X axis: Business/User Flexibility & Agility (Low to High)
This is all about those use cases that enable a business to be more dynamic so it can respond to changing circumstances faster. In other words how nimble do end user computing services need to be to meet dynamic business and user needs? For example, you may have lots of M&A activity, sophisticated disaster recovery requirements, a highly mobile workforce, or drivers to restructure your business requiring more agile end user approaches.
Top X axis: Mobility/Cloud Capability & Enablement (Low to High)
Traditionally, enterprise mobility has been about access to enterprise email and Windows apps. But we all know it’s moving beyond this fast and BYOD use cases are just the beginning. However it’s almost impossible to think of mobility alone without thinking about cloud services and how they will be consumed both inside and outside of your firewall in an increasingly mobile world.
Left Y axis: Cost (High to Low)
It’s important not to get tripped up by this one as it scales from High to Low to represent that low cost is of higher value to the customer.
Right Y axis: Well Managed and/or compliance (Low to High)
For traditional end user architectures, well managed is more desirable as it directly correlates to cost and compliance. However this axis also recognizes that with the advent of consumerization, assuming that you can manage everything is a facade. So compliance is recognized as a standalone requirement for many modern use cases.
Mapping these architectures to specific quadrants
Now that the axes are outlined, the fun begins. Based on my experience I’ve mapped various end user architectures to each quadrant as illustrated above. I struggled to come up with good summary names for each quadrant, so I did what I felt was closest to customer truth. (Again, suggestions are welcome.)
Mapping these architectures to their respective quadrants is an interesting exercise. The various axes caused me internal conflict and I had to balance what I thought was most important versus judging what was realistic. Whether you agree with my final placements below is not essential to experience the framework. Instead it's' about figuring out what’s right for you using the framework and feeling that same internal conflict. Customers get many approaches to end user computing thrown at them, so it’s worth pointing out that being placed in one quadrant versus another does not imply that something is bad or there is not a big opportunity for a particular approach. However, in an evolving world, I do believe that customers are looking more and more to move to the top right quadrant to meet their needs which, in my opinion, represents the greatest forward looking opportunity for the industry.
So let's look at the four quadrants.
I've called this quadrant the tactical quadrant because I believe the majority of enterprise customers do not broadly adopt the architectures and solutions here.
I’ve placed RDSH as the lowest cost solution because it's highly managed but offers the least amount of flexibility and mobility capability. That does not mean that RDSH is bad or not a substantial opportunity as it gets better. I just think it’s not the foundation that will take enterprise customers to the top right quadrant. XenApp improves upon this with more enterprise and cross platform capabilities. Non-persistent VDI, by virtue of a single instance desktop OS, offers more flexibility at a higher cost. I had to make a judgment call here also because I felt the cost outweighs the management, which is similar to XenApp and reflected that in the placement.
I then went through a similar assessment of MDM, MAM, email, file synch and data access approaches on the market today.
In this quadrant I started to think about the cost of third party add-ons to RDSH or VDI approaches. There are many good solutions to choose from at different price points, but I weighed the added complexity of an additional vendor, maintenance, and possibly another management layer or infrastructure to support more heavily. Of course I could have come to the conclusion that the benefits would outweigh the costs, but the truth is simpler solutions on the market today achieve that for many use cases.
I came to the conclusion once again that RDSH was the wrong foundation to make progress towards the top right quadrant. I do however believe that third party solution solutions (when used in conjunction with other architectures in the strategic quadrant) will add more value and hence should not be overlooked, although capabilities and maturity still need to evolve.
Then I took a stab at cloud aggregators. My opinion is that these are still nascent. Many customers are not ready yet, so this could just add cost. Also cloud aggregators still seem very disconnected from mobile client solutions, but they do create more flexibility to consume cloud services. I expect a lot more innovation will happen here.
Finally I considered, Enterprise MAC solutions, both existing and emerging solutions. I weighed the cost of supporting another management solution higher than the mainstream Windows PC enterprise use case and believe the adoption numbers are still too low. However as those numbers grow, I anticipate these solutions to be more strategic rather than tactical.
The reason I call this quadrant strategic is because I believe it represents what enterprises are more likely to do today when considering something strategically rather than tactically. Specifically I’ll call out persistent VDI, because people who I have seen succeed with VDI treat it as a strategic architectural approach as part of their broader IT strategy. That’s an important distinction as opposed to having frivolous debates about X% adoption rates. It’s strategic if you want to be successful.
I started with full rich clients with full user control since that's what users most likely love the most and will argue it's the easier way for them to get work done in any way the business needs. Of course this introduces a boatload of compliance and security risks. Trying to manage this type of environment is like a dog chasing its tail.
Then I thought about it in the context of better-managed enterprise PCs and enterprise MAC solutions. Even well-managed enterprise PCs still represent the challenge of distributed computing and PC lifecycle management. The best solutions here still mean slow change management in many cases, and let’s face it, application virtualization can’t exactly be described as mainstream. However, there is hope as solutions like VMware’s Mirage and other startups such as FSLogix, CloudVolumes start to attack this problem in new ways. Even from a security point of view, innovators like Bromium can make life a lot better for the Enterprise PC and beyond.
This naturally led to me to think about persistent VDI next, irrespective of deployment type (i.e., Enterprise or DaaS). I believe VDI continues to offer enterprises more flexibility than the traditional enterprise PC model or RDSH at a cost. These costs today are however being attacked mostly from a CapEx point of view. If you’re at VMworld I’d recommend you attend the @Gunnarwb session on attacking the high cost of desktop virtualization. However, I believe to move VDI into the top right quadrant. We must continue to understand why it is a better approach than RDSH for the future, and we must continue to attack OpEx as I have previously written about here, here, here and here!
Finally, since so many people I speak to tell me this is their workplace reality, I considered both collaboration and social solutions. Collaboration with things like IM, video, voice, remote meetings or Sharepoint are certainly much more established in the enterprise and many are just beginning to think about new ways to move these outside the firewall yet retain their security posture. Social, I think, is still very early days and poorly understood within the enterprise. It doesn’t have the benefit of mass participation like in the public domain and of course IT screws it up with too much control. I am by no means a social expert so throw rotten tomatoes at that one all you want.
I consider this the opportunity quadrant. I think all the various approaches today will push in this direction as they evolve in some form or the other. This presents a conundrum for enterprise customers. They will be bombarded by so many approaches that the value may be washed away with the complexity of many competing solutions.
I felt that today only two entries deserved to be in this quadrant. Email, which most people tell me is one of their most pressing concerns as a result of consumer devices and Workspot. I added Workspot because I like their approach to cloud based access to mobile devices to your existing enterprise applications. I think that is today’s mobile reality for many enterprises. Hence why I didn’t put them in the tactical quadrant like I did for MDM/MAM vendors, which currently serve a small application base and hence are of lower value to enterprise customers. I still think Workspot have work to do to meet broader enterprise needs and will have to add more client support. However, they represent the type of thinking that I believe will create new customer opportunity.
Enterprise customers will need to decide upon the types of services they offer and consume in order to determine what makes the most sense for them. Perhaps offer different service tiers for different user types. To do this they’ll need to understand a lot about what is possible and then figure out what to own vs. consume. It will be a lot more than a, “I love architecture X,” debate which for the most part I find frivolous when considered in the context of where customers are trying to go. I think the skill level and awareness of the IT world will need to be greater than it is today for IT departments to thrive in the ever emerging end user landscape.
From a vendor point of view, I think the more they do to simplify all this complexity the better. I can certainly envision more comprehensive suites that integrate many emerging approaches as the big winners with a few highly differentiated smaller players holding their own while the rest get bought or fail. For the winners, it will be a question of how well they execute and their ability to innovate to solve for new use cases. The thinking behind current mobile solutions is still very nascent and there are so many areas that could be exploited like identity and big data to make for far more comprehensive suites that take customers on a journey.
Can end user computing unite the Software Defined Datacenter vision?
Since it’s VMworld, no doubt we will have heard a lot about the software defined datacenter. Often people in the end user camp switch off and think, “Oh, we’ll switch back on when they stop talking about all that data center stuff.” However, I think the two will get closer over time. All the investment being made to consolidate, automate, and better manage the datacenter, is ultimately there to provide services to end users.
Most of what we hear is a "bottom up" approach to making this all possible. But end user computing could be the "top down" approach that unites the software defined datacenter vision and investments into a powerful end-to-end solution. Most importantly though, the winners in the end user computing space will be judged by how well they meet the enterprise customer desire of enabling a highly flexible enterprise that is well managed and/or enables them to remain highly compliant and secure, allows them to be highly mobile, and allows them to consume solutions outside and inside their firewall at a low cost in a simple way. It’s an exciting time ahead for end user computing!