What Makes VDI Insanely Great? (Hint: It's Industry Chops)

Author: Brad Peterson

Publish on: Sep 21, 2016 6:00:00 AM

This is the second article of our series on the big picture of VDI and VDI 2.0. If you've been reading our posts of late, you can tell we're thrilled about the enthusiastic response we're getting at shows like VMWorld (where we won Best In Show) and Nutanix Next. Because we can't meet all of our readers and friends at live events, we put together a series of videos that enable us tell the Workspot + VDI 2.0 story. Below is the transcript of video #2.  (Here is a link if you would rather watch the video.

Amitabh: So we started Workspot nearly four years ago in the summer of 2012.

Brad: No wait a minute, Amitabh. This story starts much earlier than that. Let’s start earlier -- in the days when we were at Citrix.

Amitabh: Ok. I joined Citrix in 2006. I spent the first few years building the first version of its receiver Project Pictor. If you remember, that was the big graphics project back in the days when GPUs were not part of the data center. After that, I joined the XenDesktop team as Vice President of Product Management. At that time, there was zero million in revenue. By the time I left it had earned a billion dollars in six quarters. After that, I was a General Manager for ZM and XenDesktop.

What I found problematic at that time -- and the reason I ultimately left Citrix -- was that all of these customers were buying XenApp and XenDesktop licenses and VDI licenses, but very few of them were able to deploy it. It was too expensive and too complex. All the problems were in the data center. So the data center architecture at that time was servers and storage. You bought NAS or SAN from EMC and storage from Dell, and you plugged it all together. And it was really not built for VDI. As a result, Citrix was running into all kinds of problems. Projects were taking months -- sometimes years if they failed. It just didn’t seem right, and so I left.

I just knew there had to be a better way for doing VDI. It had to be simpler.


Brad: Did you (or Citrix) know a better way at that time?

Amitabh: No. I didn’t know of a better way. My thinking was that to get VDI right, you had to start with a clean sheet of paper. So my first step was that I needed to figure out what a better way to do VDI actually was. And then I could go design that.

Brad:  At that point you go off and create a company from nothing.  Tell me what that experience was like.

Amitabh: Our philosophy behind starting Workspot was answering the question: How do we make it simple for end users? How do we make it simple for IT?

So we decided on the parameters -- on how to make it simple, which is how do we do it? We leverage everything that exists in the data centers so that IT doesn’t have to install boxes for 9 months.

How do you use the power of the client? Today phones are more powerful. PCs and Macs are more powerful. How do you use that power to your advantage? Also, the cloud is now available. How do you leverage the cloud? And so when you put all these things together, you can actually design a completely different architecture that is simply not possibleto create with the old legacy architecture in the data center.

Brad: So a team is required for that.

Amitabh:  Workspot’s CTO and co-founder Puneet came from VMware. He was actually the first engineer working on VDI in the world. He built the VMware View product. And he left VMware for the same reasons I left Citrix-- the product was too complex. We bought in Ty Wang, who is another Workspot co-founder. Prior to Workspot he was at a company called Twilio, which is the number one company in the cloud based communications industry right now. They just went public, actually. And so we brought our initial team together, and then we started recruiting additional team members.

We wanted client engineers who understood the data center really well, so our first two client engineers came from Cisco and Juniper. They know how to build a client that leverages the power of the client and connects it back into a Cisco and Juniper box.  We brought in cloud analytics people because we wanted to use the cloud to capture a lot of data. Those people know how to analyze terabytes and petabytes of data. Actually, that DNA didn’t even exist at the time, so we had to assemble it out of people who didn’t know VDI and didn’t know app security.

These were people from very different industries. But they knew networking, big data, clients, and the client cloud all really well.

Brad:  You assembled the team at Workspot and started building it over two years. Meanwhile all this is going on, what’s going on at Citrix? 

Amitabh:   So just as we were building the first big XenDesktop release (the FMA release) -- rolling it out -- there was a lot of discussion going on internally about what to do about the cloud. I gave people the example of Siebel versus Salesforce CRM 1.0 and CRM 2.0. [You have to] start from scratch for the cloud -- you can’t just repurpose what you already have. And there was a lot of discussion going on. And people said: “We want something quick, so let's host it in the cloud.”

And I completely disagreed with that because every single legacy software in the Valley has made the same mistake. There is no shortcut to the cloud. You [have to] start from scratch. Build it from scratch. Salesforce did it; Netsuite did it. There are all kinds of examples of companies that first built the product from scratch. But [they] also built the business model from scratch. Right? Because you can’t go from a perpetual business model to a subscription business model without the right product behind it.

Brad: Very few companies have done it successfully.

Amitabh: That’s right.


Brad: Adobe mad the shift. So did Office 365 and Exchange. There are a couple of examples. But very few companies can shift to that sort of movement. So you went off and built the company. You’re two/three years into it now, and you’ve build VDI 2.0. So that’s when the game changes. Tell me more about that.

Amitabh: So the first two years took us time to build the client pieces out and the cloud pieces out. And [during that time] what we realised had happened in the industry was HCI and flash had come along and solved a lot of the data center problems, right? So flash made storage fast, and HCI simplified the whole data center architecture. Simpler so you plug-in to get all this together. The data center architecture now has become 10 times simpler. It's built for VDI essentially. And so we took our client and cloud and said: “How do we use the power of this new data center architecture?”

And that’s really VDI 2.0, which is the data center stack is simpler. We took the entire orchestration layer of how do you create desktops, how do you create secure apps, how do you do policies, how do you updates plug the two together? And that’s VDI 2.0. It’s dramatically simpler than what we had built at Citrix six years before. And a lot of it is attributable to the fact that cloud didn’t exist in 2009. HCI didn’t exist. Flash was just coming on the scene at that time. We had the advantage of starting three years, four years later. And then the data center architecture completely changed just as we were getting started. So VDI 2.0 is only possible now, almost, in 2015/16.

Brad: With Workspot’s cloud architecture combined with HCI and how it’s elastic, right? The fact that you can get VDI users up in a day -- as opposed to weeks and months, and so forth. You know, that’s dramatically different, right?

Amitabh: If you think back to VDI 1.0 projects, people would be happy if it got done in a year. We are not happy if it takes more than a day, right? Our goal was answering the question: How do we get it to be really fast and really simple? We have a challenge on our website, right? We will get you up and running in 60 minutes. So that was a focus -- forget months. We want to do it in minutes.HCI__Workspot.png

Brad: So where is the target market? Where is the sweet spot? Does VDI 2.0 go small-to-medium? Mid-market? Enterprise? Where does it fit?

Amitabh:  The surprising thing is right now we have three different things happening all at the same time. So number one is all the HCI vendors, so thanks to Nutanix, Simplivity, Atlantis computing, Scale Computing, Cisco, HPE…. They’re very excited about their hyper converged appliances. They are also very excited about VDI 2.0 because it helps them sell their boxes faster. From all across the world we are actually getting in-bounds from these HCI vendors saying: “Hey, I want to be a Workspot partner because it will help me sell my HCI boxes faster.” Those guys are mostly going after the mid-market sector.


And the second thing that’s happening is that you and I know that all these large enterprise customers of Citrix, you and I know a whole bunch of them right.

Brad: Yep and from EBC.

Amitabh: They all know us. We understand their markets really well. So we are getting in-bounds from them. I mean, you know the folks we are talking about right now. They are all struggling with XenDesktop right now. And they don’t want to expand it or do a new project with XenDesktop. As a result, they are talking to us.

The last one, which is a little bit surprising to me, is how much the resellers want to become MSP providers. They all need to figure out what is happening with the cloud. How can they survive in a world in which apps are all moving to the cloud? VDI is a perfect solution for them. It helps them get closer to the users, closer to the customers, and enables them to provide a mission-critical service. So we have a whole bunch of re-sellers who are interested in becoming MSP providers. So all those things happening simultaneously. The MSP’s are mostly going after the SMB Market. The VARs are mostly going after mid-market. And the large customers right now, we are not even going after them ourselves. They are coming in in-bound right now.

Brad: Well, with this unlimited potential, you know, it is interesting right now. It's very exciting times. What is the big goal for Workspot? Where is the future? What’s the opportunity?


Amitabh:  We want to do a million users on our platform. We want to be the first platform to have a million users running desktops and apps, right?

That is a big deal. We architected the platform exactly for that, right? So it’s built to be highly scalable. It’s cloud native. It’s microservices. [When I say] infinitely scalable, I mean that [it is] as scalable as Amazon is right now. And it’s highly available, [which means that] we designed it so that the cloud is never a single point of failure. Because that is what freaks people out, right? If you are a broker and you become a single point of failure…. So it’s both scalable and more available than VDI 1.0 architectures.

The other very interesting thing that we did was we instrumented the client from day 1 to collect data, right? So we collect data about user behaviour: What apps did Brad open? How long did it take for Brad to access apps? What devices did he use? Where did he use it from? What network? AT&T or TMobile or home wi-fi? We are tracking everything from a user behavior perspective and sending it up to the cloud. Think of it: you have a millions users, thousands of customers.

And we have the data across all of these customers, across all of these users. Real time performance data [that’s transmitted] back and forth. Real time security events [that are transmitted] back and forth. And we all have the tools in place [to enable you] to analyse the data and make IT more pro-active. With VDI 2.0 you aren’t waiting for the phone to ring with 200 people saying: “You know what? Your SAP application is down, we will tell you all about it.” We are not waiting for security teams to be reacting to events happening in the field. We will give you real time data about who is doing what, from where, in the field, so you can plug it into your SIMs systems (Splunk, for example), analyze the data, and react to the data. That’s the big goal…. We have all the pieces in place. It’s how we’ll get to one million users.


Brad: Right. So how does it make you feel being in a position where you are now? Compared to where you were years go at Citrix, doing such a volume business but struggling because the architecture wouldn’t allow it to be successful? And now you are where you are.

Amitabh: You know, yeah. We struggled a lot with the VDI 1.0 deployments. Part of it was us, but part of it was the data center architecture, right? And both put together just was a disaster for customers. Now our architecture is really scalable. And we are now actually doing POC in boxes with customers where we give them 500 users for 30 days.

That’s the power of the platform we have now.




Want to learn more about the power of this platform?  Check out our eBook that outlines just how could implement VDI 2.0 in the time you eat your lunch. 

New Call-to-action



Subscribe To Our Blog

Recent Posts