So I’ve been conducting a number of studies recently and a client asked me why in a recent study I hadn’t considered tablet’s as a PC replacement (primary device) when looking at VDI DaaS solution. When the question was posed it was something that I had actually not given a lot of thought, luckily with a brain full of VDI knowledge I worked in in nanoseconds that replacing my thin client with a tablet (non Full windows device)essentially had the same effect (just with now a mobile management effort required as well). Now having had some more time to ponder this in detail I thought I would give some further insight (any tablet donation’s are welcome!)
VDI = Reduced TCO???
So without going into all the details it’s generally a widely known fact that the direct cost of VDI is greater than traditional architecture (for those who dis believe this the main components of additional cost are VDA licensing, VDI broker/management licensing and server/DaaS cost).
The saving’s we make on thin clients/windows thin over a period even both those options out and the elements above overall make PC architecture generally cheaper.
So let’s look at using a tablet
Ok so I’m going to replace my thin client device with a tablet device which is not Windows Logo certified. For this I’ve picked the Sony Xperia Tablet Z for pricing.
So as we can see it’s probably a few pounds more expensive than a thin client (I usually ballpark about £200 for thin client), it does still require VDA licensing if we want to provide the user a Windows Client based VDI session. We also need to manage this device using a mobile device management system. So already we can see that it’s cost is greater than a thin client and therefore greater than traditional architecture. It is however a mobile device with offline functionality.
Now we still need to think about security, we need to ensure the device is encrypted and due to the lack of reliable android antivirus solutions we need to lock down the apps that can be used by the device, enforce security unlock and have remote kill etc. etc.
Now since I’m a Windows man and I don’t have an android based tablet I’m going to have to call out that what I’m saying is theoretical, I have however just downloaded Genty Motion (https://cloud.genymotion.com) so that I can at least be a bit more accurate with my statements. I’ve even used the Xperia Image.
To replace a PC with a tablet I’m going to try and not limit the device to consumer only users
So now we have a scenario where we have a tablet, we now want to give our user the ability to actually create content. Using my experience with my Ipad and Surface RT the use of touch alone left me without a warm and fuzzy feeling, on my ipad as a replacement to my laptop for a week project I failed on day 1.
So first thing is first we need a keyboard! I’ve seen a range of these available and my friends uses one for his Samsung Galaxy tab. I’m going to assume we get some track pad functionality with our keyboard.
So out of the box (plus a keyboard/trackpad later) we now have a device that we can attach to our corporate MDM solution (we’re using corporate WiFi).
Over the wire configuration
So the next step I’m taking is to connect to Office 365 to let my activesync policy kick in. I’ve not been able to use auto discover, so I had to go to the Office 365 site to verify the correct configuration.
Now we need to be productive (offline), so I hunt down a free Office App (this seems to be easier on a physical device!) as I have to install the google+ apps. But without too much drama (a few error messages). I’ve got my device syncing email/complying to the company security policy and after several crashes I think I’m ready to get a productivity and some VDI/RDS/Citrix apps installed. Just a quick note, it seems RDP without a third part product will be a requirement for this experiment (=additional cost).
So the two main options we have are: VMware view and Citrix.
From a cost point of view I’m going to use my previous research to make some assumptions here. The 100$ per year VDA license is prohibitive, so I’m going to say we have two options here, one if to use XenApp, the other would be to negate VDA by using Server 2012 as the desktop, both of which may have potential issues with line of business applications from a support and/or licensing point of view. Now VMware doesn’t support Server 2012 as a guest OS however Citrix does. This now goes down to the question of do we really need to provide a full desktop to support the line of business applications and productivity set. We’re back in the same VDI situation however we look at it from this point of view. From a cost and support point of view using XenApp to publish the required apps (using app virtualisation where required) seems like the best option if we need to present a windows environment to the user.
It’s all about the applications
So if we were actually going to venture down a tablet based route and we accepted the costs of managing, supporting and maintaining the associated systems what would the best route be?
To my mind the current utopia would be to deliver the app via HTTPs or a device compatible app. The cost in delivering your current systems (and if the interface changes would in fact be sensible) to a tablet based device would need to be assessed, it’s highly likely your line of business applications currently are not in a format whereby there is an android based application, the cost of development/upgrade may be extensive.
In the future in the enterprise this may be a very viable option, for now though I can only see that taking this new tablet consumer trend and applying it to the enterprise world is going to be a bumpy and costly ride when compared to the traditional PC.
The death of the PC… I think not, the considerable reduction in volume sales.. well we can see that’s already occurred. For me, I’ll be sticking to my Windows based desktops/laptops and hybrid devices for the time being where business is concerned until application vendors can show a like for like experience on both formats.