commentary In 2008 virtual desktops were one of the hottest IT infrastructure topics, however, the tightening of IT budgets from the economic downturn, and mounting evidence that virtual desktops are more expensive than well-managed full desktops, will dampen enthusiasm for this technology in 2009.
IBRS advisor, Kevin McIsaac
(Credit: IBRS)
This was clearly highlighted at a recent peer network round-table, which we facilitated, where we confirmed our long-held view that virtual desktops are not a general purpose replacement for a full desktop PC and that reports of mass roll-outs of virtual desktops are pure vendor hype.
The peer group at these round-tables consisted of desktop and infrastructure experts from 19 Australian organisations, many of which are the largest IT-consuming organisations in Australia and New Zealand.
As part of the discussion, we polled the attendees on the number of virtual desktops deployed into production and found only five (26 per cent) responses.
Furthermore, the two largest production implementations had only 700 and 600 virtual desktops respectively. Out of the approximately 286,000 desktops deployed across all 19 organisations, only 1,585 (0.55 per cent) were virtual desktops and the highest virtual desktop penetration in any single organisation was only 2.3 per cent.
Based on the plans that these organisations have to roll out virtual desktops, a generous estimate is that by the end of 2009 production of virtual desktops will be 1.6 per cent of all desktops. A more realistic estimate is well less than 1 per cent.
What is now clearly revealed, is that even after years of intense marketing, extensive product development, and in the face of considerable interest from IT organisations, the uptake of virtual desktops is still very low and mainstream adoption is a long way off, if ever!
This reality should be compared to vendors' press releases, and the accompanying news stories, that routinely indicate massive uptake of virtual desktops and position it as a revolutionary technology that is on the verge of going mainstream.
In our research we identify five common scenarios for using thin desktops:
WAN-enabled client/server applications: most of the organisations we spoke to already used Terminal Services or Citrix Presentation Server for delivering some applications. Since Citrix is highly effective in this scenario, and current virtual desktop products do not support delivery of individual applications, it is unlikely virtual desktop use will grow in this scenario in the next few years.
Small remote offices: a few of the organisations were using Citrix Published Desktops for this purpose and there was some interest in using virtual desktops in this scenario. However, there was no clear consensus at the round-tables that a virtual desktop would work better than Terminal Services or a Citrix Published Desktop.
Of the five organisations with production implementations of virtual desktops, all were used for offshore developers.
There was considerable discussion about the impact that introducing thin terminals has on supporting new technologies, for example VoIP, multimedia and unified communications, with the view that thin terminals were currently constrained, but that this constraint was slowly being lifted.
Secure remote access: several organisations already used Terminal Services, or Citrix, to enable users to access a desktop from remote locations. At least one organisation had optimised this to enable staff to seamlessly move from a full desktop in the office to a thin desktop at home, while retaining the same user experience — that is, access to data and applications.
Over the next 18 months, as new technology enables virtual desktops to be migrated from the server to the desktop and back again, the use of virtual desktops may start to challenge Terminal Services in this area.
Harsh environments: while there was some discussion about this, there were no clear consensus about the benefits of using thin terminals over PCs in a harsh environment.
BPO and offshore developer desktops: interestingly, of the five organisations with production implementations of virtual desktops, all were used for offshore developers. As discussed in earlier research, this is the best, and perhaps only viable, scenario for virtual desktops.
While there was interest in the use of Virtual Desktops, all organisations found it difficult to create a compelling business case that favoured a virtual desktop over a full desktop, Terminal Services or Citrix Application Delivery technology, except for the BPO and offshore developer desktops niche.
Dr Kevin McIsaac is an advisor with Australian analyst firm Intelligent Business Research Services. He has 25 years of IT experience and is a recognised expert in infrastructure, operations and vendor management. Prior to his current role, he was research director of META Group's Asia-Pacific Group division. He has also held leadership positions at Computer Associates and Functional Software.
2%
4%
The author missed one important point, virtual desktops comprise terminal services, VDI (virtual workstations on hypervisors, VDI) and PC blades (physical hardware, DDI). VDI isn't the only game out there and you also have to include technologies like Parallels Virtuozzo.
In addition the advent of multi-core CPUs, VM provisioning and linked clones means the cost of VDI is dropping significantly.
Poorly researched, poorly defined.