A great Aussie virtual desktop case study

7

rmit

blog Virtual desktops, bring your own device computing, integrated datacentre components. These are three of the hottest trends to hit Australia’s enterprise IT sector at the moment, and they all come together in this highly recommended article by iTNews writer Chris Jager looking at a huge virtual desktop implementation at RMIT University. A sample paragraph, quoting RMIT IT services chief Brian Clark:

“As far as I know, we have the largest BYOD VDI environment in Australia, so I’m quite proud that we’ve got what is a very technically complex environment running and performing well at this scale and capacity.”

We’ve been seeing the building blocks of this kind of rollout being put in place in Australia for a while, but few have put those pieces together into the kind of integrated platform which RMIT has. I suspect that, just as web-based cloud email has taken Australia’s university sector by storm over the next couple of years, this kind of virtual desktop rollout may be next. Highly interesting stuff.

Image credit: RMIT

7 COMMENTS

  1. Half of me thinks BYOD is a disaster waiting to happen, the other half thinks it’s the best thing to happen to IT since the GUI. Virtualised desktops/environments are great, if configured properly, ie: security is hardened & access is easy for the user.

    I’ve pushed for this kind of set up in many work environments, however it’s hard to convince old school admins/architects that moving to a virtualised environment is a smart thing.

    Good to see it picking up steam.

  2. Shame they didn’t cover more on how they got around all the licensing issues the various vendors have for when their software is run in a virtualised environment.

    • Licensing? *cough* education *cough*

      You do know that Google apps/mail/drive (as an example) is free (as in beer) for education, right?

  3. I’m still VERY much looking forward to the day of completely portable desktops. You grab your tablet, pop it into a dock and it becomes a full blown, hard core desktop, pulling all your settings, programs, games, everything from your home or work server.

    I know many virtualization systems can do this. I’m just looking forward to it being truly portable and on any device. RMIT seem to be plowing such ideas forward :-)

    • I can see the appeal, but I think maybe the portable go-anywhere solution is coming from the other direction. Later this year we’ll see Haswell, which will have tremendous performance on tap even for low power computing. With full native integration of Thunderbolt, we might finally start seeing some decent connectivity options, allowing 1.3kg ultrabooks to connect up to external graphics with desktop-equivalent performance, tripple monitors and all the storage a desktop could need while providing more than enough CPU grunt for all but the most demanding situations, and yet is still so unobtrusive it can be slipped into a briefcase or satchel to be taken to meetings or on the train or wherever. I see this approach evolving over the next few years to tighten up voltage leaking, strengthen power gates and completely disconnect components from power when they’re not in use. This could result in tripple or greater running time on exactly the same batteries as in use today.

      • @Trevor

        Oh for sure. Has well and evolutionary architecture is REALLY exciting. The idea of having a sub 1kg Ultrabook that acts as an internet browser, document editor and TV watcher on the move and become a full boar video editing suite when you get home practically has me salivating. I much prefer physical keyboards (he says typing this from a Nexus 7….) and I would certainly be one of the ones on an Ultrabook rather than a tablet.

        I only mentioned tablet because, for many people, that’s all they need and want. But they want more when they get home. I think we’re gonna see more and more blurring the lines of PC and portable devices as we go forward. I think it’ll becoming device in dependent how we compute- you could literally have a tablet acting as your home server when you plug it into its dock at home. Or your Ultrabooks. Or you phone. Or whatever way you prefer to have portability. It’s clichéd but it’s exciting times we live in. :-)

  4. Sorry, I actually fell asleep writing that, so I didn’t get to complete my thought *looks sheepish*

    Yes, on the flip-side you have low power, cheap tablets (that have essentially replaced the netbook market) that do have a lot of potential via virtualisation to be full-functioning work desktop portals. The NBN will play a pivotal role in enabling this changing landscape, as currently experiences differ wildly outside the LAN, making provision of full desktop experiences ubiquitious across the organisation problematic.

    Having powerful cloud services bundled with advertising or simply provisioned via a subscription model will allow prividers to experiment with profitable models delivering a similar environment to consumers on ‘thin’ devices, which I do believe will work well for a very significant proportion of the population.

    But there will always be tasks that don’t lend themselves well to cloud operation. Today that includes transfer, manipulation and use of large files such as video editing and compressed file packages, tomorrow it will likely involve 3D models for 3D printing, and probably a whole lot of things we haven’t even thought of. I think complex encryption to safeguard privacy will also become a mainstream practice, to allow people to store their lives on the cloud without fear of interception or snooping, something people will only be able to do locally (as encryption via an unknown host is pointless).

    Thus I see the future as a convergence between these two approaches, with no real liklihood for dominance by either, as both approaches are valid and provide necessary features. The reductionist view that computing will move pervasively in one direction is narrow-minded and naive – there will always be both those who want computing to be cheap and as simple as possible, as well as those who need as much performance as possible and a flexible, adaptable platform that can be customised to support the tasks they need to perform, and they will be prepared to pay what’s necessary to achieve that.

    Will we ever see a ‘post-PC’ era? Perhaps, one day, if network technologies manage a quantum leap in performance relative to storage capacities and user models, and cloud (ie server) processing can achieve a per-user cost/performance ratio that exceeds that of PCs and workstations. At that point there will essentially be no point in owning PCs that sit under your desk for local processing power or faster/cheaper file storage and retrieval when a thin client and cloud services can do it faster and cheaper anyway.

    But we’re not there yet, not by a long shot. Yes, local clients need to be adapted to be far more power efficient than they are currently – it is ridiculous for a PC to be drawing three or four hundred Watts idling at the Windows desktop when there are ultra books running the same OS that can do it for 20W. The lack of Thunderbolt options right now is likewise limiting what might otherwise be a revolution in the way we see and use our computing environments. So I look forward to these and other innovations that will change the way we see and use computers, but am under no illusion that these changes will see the ‘death’ of powerful, flexible client machines in the foreseeable future.

Comments are closed.