I thought I’d jot down a few notes around the good use cases for VDI. I often hear people sell the idea with a reduction in cost as a key driver. It’s true you can use the VDI initiative to reduce the direct total cost of ownership, but in reality the VDI element is not what causes the reduction in TCO. (I am also assuming we are reducing TCO based on a BASIC level maturity orginisation).
So without using reduced TCO as a driver, when would I recommend VDI?
Below are some high level descriptions of scenarios where VDI could be a good fit:
- Job Roles that require no offline productivity or offline LOB application access (remember some systems use parent/child database replication)
- Users that only roam to internal fixed locations (e.g. Receptionist from one front desk to another)
- There are no physical ties to local services (e.g. serial cable into a waiting room display system)
- Printing is not conducted in a high volume/high bandwidth use pattern (e.g. Plotting AutoCAD file’s over the WAN may not be a good idea)
- Users system access is bridged (e.g. the LOB application platform provides a presentation layer within the same physical domain and has only a limited surface area of access)
- No hands and eyes are available at the users location and to do so would be cost prohibitive
- The centralization of services is key due to regulatory/company policies
- When an additional access layer is given to allow multiple device support (when the primary device not require VDA) and productivity benefit is deemed to outweigh cost
- There is a requirement to provision a large number of devices from a well connected location in a very short time period where a standard configuration is required.
- Providing contractors/3rd parties system (Desktop) access when providing VDA/VDI services is not cost prohibitive
In addition to this the use of VDI to cater for disaster recovery scenarios can greatly reduce the effort/cost required to maintain a DR site’s client access services (thought the same could be said about server based computing)
For call center, kiosk based and task-driven worker roles VDI can be a great solution if your existing architecture doesn’t have a suitable mechanism to cater for this scenario. If your driver is not cost saving and is based upon speed of delivery and/or task based worker scenarios then VDI has the tools to assist in these scenarios. But so does RDS/Citrix, web applications, VPN’s and a number of other technologies. To me I would unless fitting into a specific business strategy where it is clear the benefits of VDI outweigh traditional architecture (e.g. rapid provision/de-provision of sites), I would question why you can’t achieve the same results without the additional cost.
Years ago we used to have a desktop and then we could request a Citrix based remote desktop environment (at additional recharge cost, using folder redirection, loopback and if required WOL so that you could RDP to your full desktop). This worked then and I see no reason why this doesn’t work now, in my mind it suits far more uses cases than the VDI desktop based solution.