With businesses and the IT functions that support them working harder than ever to trim inefficiencies, headcounts, and costs, the future looks bright for thin computing. But it did last time. And the time before that...

As Russell Brand and Jonathan Ross are no doubt discovering as they stumble sheepishly into 2009, some things are mightily difficult to put behind you. Indeed, it is those episodes that we try so hard to consign to our pasts that so often end up having the greatest impact on our futures.

So is it with thin computing; a subject whose fortunes moving forward are impossible to talk about without reference to its somewhat stop-start history. Thin Computing has always flown well as an idea, but while it never had too many problems getting off the ground, staying there and gaining altitude have proved rather trickier.

Indeed, the industry has been predicting the worldwide domination of the thin client for rather longer than some of the aging hacks on this publication have been hacks. “The era of the PC is almost over, and the era of the thin client is about to begin”, said Oracle CEO and industry seer Larry Ellison, in the mid 1990s. Not to worry Larry, only about a decade out.

Instead, while the PC has continued to dominate, the thin client has taken only a relatively small share of the market, winning some minor skirmishes but few of the major battles. According to many however, with economic factors and emerging technologies like virtualisation pushing thin clients into the mainstream and businesses beginning to see tangible benefits from thin roll outs, its future looks quite different this time round. 

Some, like Stephen Yeo, strategic marketing director at Igel, believe that thin is about to take off in a big way. Companies will tail off their usage of desktop PCs in the near future, giving laptops to their mobile workers and thin devices accessing virtual machines to everyone else. Driven by virtualisation why wouldn’t they? It’s making a big noise everywhere else, why not  the desktop?

Others, analyst IDC among them, are a little more circumspect, citing resistance from a PC-focused IT staff and corporate knowledge workers and the still massive installed base of the PC among the potential obstacles to fast, widespread future thin client adoption.

People still keep a lot of confidential and personal information on their desktop machines and will likely resist any move that forces them to store such data server-side while the 300 million PCs shipped in 2008 dwarves the 3.7 million boasted by thin client vendors. Accordingly, and perhaps justifiably, many organisations still see a shift away from the traditional desktops as risky.

But Malcolm Wilkes, strategic director at Avanquest thinks it’s now merely a matter of time and the successful continuation of the main technology bell-curve before the thin client reaches critical mass. “Providing the datacentre infrastructures can support the growth and no disasters affect the availability and speed of the Internet, then confidence and acceptability will grow.” 

Confidence and acceptance will also continue to be driven by factors such as the emergence of VDI, says Simon Ponsford, CEO of Cranberry – thanks in large part to its greater flexibility compared with traditional server based computing platforms. For now though, he only sees VDI being implemented in larger corporate environments.

“For the rest of us there will be an emergence of small footprint devices with local disk cache (probably solid state)”, he says. “Operating systems will be virtualised and delivered over the LAN, WAN or Internet and be cached locally (and) updated with new applications when connected to the network.”

“The future is a virtualised OS running virtualised applications delivered on a subscription basis to small powerful devices that have a local OS just sufficient to download a VM from the network and cache it. In the event of an application issue, it will be reset, if a virtual OS is damaged it will be patched or downloaded again.”

As Netvoyager CEO Jamil Aboulzelof points out however, the slimming down of the hardware is only half the story. For the future to pan out as the thin protagonists hope, application design has to evolve too. 
 
“It’s all well and good talking about physical hardware virtualisation using VDI et al, but the success of the thin client computing model is the evolution of how applications are written and delivered”, he argues. “At present, most commercial applications are monolithic, not cross-platform capable and understandably so.”

He sees technologies being developed by “trendsetters” like Microsoft and particularly Google, as especially key to the future of thin computing. Technologies “that will allow applications to be developed in a more modular structure; cross-platform shared libraries enabling applications to operate on any operating system independently or within delivery mechanisms like web browsers.”

Here, he says, thin client vendors will have to cater for all means of application delivery as no two end-users are the same. So while traditional screen-scraping technologies such as Microsoft Terminal Services, Citrix XenApp, and NoMachine are supported, vendors will have to ensure their embedded browsers support the real-world applications now being developed in Java, Ajax, and Google Docs.

“Applications have a re-development cycle averaging about 5 years, which means over the next 10 years, we’ll see more applications (commercial or internal custom ones) becoming ready for any platform, anywhere.”

Looking further ahead, some commentators think the thin client could even (eventually) find its way into the home “... with PC-like functions arriving via TV set-top boxes or through devices that plug into the back of televisions. ISPs have already started exploring the idea of selling such products and monthly subscriptions to computing services.”

All looks rosy in the thin computing garden then. Whether, as many predict, it will thrive in the impending cold economic winter, as yet remains to be seen.

Just a little bit of history repeating

The following is an extract from a 1998 article in ENT magazine on “the future of thin client/server computing”. Ring any bells?

“The thin client/server computing architecture is making a comeback of sorts. With the release of Microsoft Corp’s Windows NT server 4.0, Terminal Server Edition, the half-dozen Windows-based terminals that support it, and the emergence of Java-based clients, the resurgence seems well on its way...”

Thin set to wax fat

Moving forward, thin computing represents a viable option for a whole spectrum of enterprises, with IDC predicting annual growth rate in thin client sales of 22% until 2010.

This looks to be being driven by a number of factors – adoption hurdles being negotiated and or negated, application issues being remedied, performance being enhanced; the likes of Microsoft and Citrix making good headway in improving application compatibility.

The decrease in virtualisation pricing and the amendment of vendor software licences to support virtualisation has also helped – bringing it within budgets and the reach of medium-sized and even small businesses.

As such, with the service delivery capabilities of thin client and server-based computing platforms moving closer to the PC environments businesses and users are accustomed to, the way forward could now be clear for thin to win.