Monday, February 20, 2012

BYOD: a failure waiting to happen...

Mid to late 2011, a new buzz word was running rampart in many corporate IT departments: BYOD (bring your own device).  Basically what this trend centers around is the idea that users would bring their own device be it a smartphone, tablet or laptop from home and be able to use it within their work's IT infrastructure and environment.  To me this feels like a trend we saw gain popularity but then pull an epic fail.  Keep on reading for my reasons on why BYOD is just that, a buzz word.

Back a few years ago around 2007, the big buzz word back then was "green computing". The underlying premise behind green computing was to basically make using computers more energy efficient and renewable.  The biggest ways most laptop makers at the time tried to sell consumers on this idea was by boasting ULV based processors.  ULV = Ultra low voltage.  Their reasoning was that by reducing overall voltage used by a computer, it would in turn use less power and be more energy efficient.  They were right, ULV chips definitely use less power than most non ULV based cpus; however they missed one huge thing:  terrible performance.   The ULV chips have matured quite a bit since then but still are not a very popular cpu to be used in a work environment since you need to get stuff done more than you need to cut back on a minor energy savings.  So in my opinion, green computing has failed for the most part.

Its been a success in bringing awareness but its practical uses and applications in work environments are limited.  On top of that, the recession basically countered any progress it had made due to budget constraints.  Lets jump back to today, 2012.

The new buzz word floating around most corporate / work environments is the idea of BYOD (bring your own device). CIO and IT Directors are being pressured by higher level users and/or ceo types on why xxx company doesn't use yyyy type of laptop or phones or tablets.  Bascially, we are seeing consumer technology catch up and eclipse business tech due to the tech replacement cycles in work / corporations to be on a 3 - 4 year cycle.  A lot has happened in the last 4 years!

Most companies that are considering such a program with users believe that they are saving the company spent on hardware (laptops, parts, phones, etc) while still appeasing the requests of their user base.  I can understand the philosophy behind this rational and agree that initially it can save money for a company when looking merely at hardware costs/savings.  The big problems with BYOD are not obvious if you don't work directly in Desktop / Application deployemnt and Tech Support.   From the outside it appears to be a win-win situation: users get to use their technology of choice and the company saves a ton of money on hardware.  However.. there's much much more to this story than the positives; I'll  outline a few of my personal biggest problems with BYOD approach to IT.

1. You don't save that much money!

Sure, you don't have to buy hardware but you do need to now create or implement a virtualized desktop type of environment in order to support multiple OS and hardware type.  However to support your full user base, you will most likely need to setup a clustered data environment with multiple virtual servers to run the desktop applications.  Server racks or UCS for this type of setup are very expensive and require a ton of memory and cpus.  Also, you would still need to purchase licenses for the software being utilized in the virtual desktop and also would need  to pay for the virtual desktop OS (be it a thin client or virtual windows machine).   Plus, you can't forget about support contracts for the new server hardware, costs of getting environment setup and most likely the hiring of several virtual desktop administrators.

2. Performance or choice?

We have all seen this trend hit before however back then it was mainframes and terminals back in the late 60s to mid 80s.   The one problem you have with a mainframe/terminal setup; the more users on it the greater degradation of speed and responsiveness for each user.  The main supporters of BYOD believe that they would in turn use their super powerful personal laptop at work and in turn get the benefits of using a power pc vs their older work laptop. WRONG! Local power has little relevancy in a thin client/virtual desktop.  You meet a basic requirement for the standard virtual desktop and anything above would perhaps help a little but again the processing is done on the server end of things.   One way around this would be to create a "live" network boot os that workstations would boot to and then run on their laptops; this would actually utilize their local hardware resources and in turn give them benefits of having a powerful laptop.  I think we'll see the "live" network boot approach this year, it seems like a very good compromise!

However one good thing about setting up a virtual desktop is that you can also support the tablet devices given there is an app/client for the iPad or Android based tablet.

3. Compliancy / Security.

Users don't like these two words at all.  To them they see "reasons why i can't bittorrent at work or another password to type in".  Us in IT see this as, "making sure our company's assetts aren't stolen or land in the wrong hands; ie keep everyone with a job".  Bringing in of personal devices makes enforcing password or encryption policies even harder.  IT fights with its users enough when the devices are company owned, now think about that if the devices are the users' own?  Exactly.

4. SLA / Expectations.

This is where BYOD really can get ugly if you don't put in company wide policies or even if you do.  SLAs and expectations need to be set asap if you do go BYOD.  Anything that goes wrong with the hardware will now be considered "IT's fault" and any problem (including the why can't i upload pics to facebook) will fall into IT's realm.  I get at least three questions a week about a personal/home pc at my work.  I actually had to create a side company to handle these requests (I'm just too nice and will fix anyone's pc).

5. Growth and training

If a company does go down this route, they have to realistically look in the mirror and man/woman up.  You can't expect your current IT staff and environment to support this environment; you need more people and training.  These technologies are fairly new and still not without faults, bugs.  IT departments need to start thinking about support personnel numbers in comparison to devices/servers/technologies utilized instead of the traditional users to tech staff ratios. Traditionally, companies didn't run too many servers due to having to have physical space, cooling, etc for them; however 2012 virtual servers are the new trend.  A company that ran 30 physical servers can easily run 100 virtual servers and since they are virtual management has no idea that the workload for the IT staff has basically increased by 3x.

Also, virtualization of desktops requires the apps to be virtualized and migrated.  Users need to get trained on new setup and security needs to be audited too.

So there you go, my reason why I don't like BYOD.  Its a great idea and can work in most cases but companies looking to go this route really need to look in the mirror and ask the big questions.  Are we really saving money?  Is it worth it?

I work in IT and see the next 5 years in IT support as being some of the most challenging years yet.  Your thoughts?