As organizations are planning the implementation of a virtualization strategy, they often forget something that’s very important — technical support. Technical support is an often under-discussed aspect of technology acquisition, though is part of nearly every technology-based solution. Depending upon the type of software involved, technical support contracts often cost between 20% and 25% of the software license fee for each product that makes up a solution each and every year that solution is in use. Organizations are often not clear on what will happen to supplier’s technical support when virtualization technology is added to the equation. Does this problem show up on the radar screen in your organization?
What's the problem?Applications are typically layers of technology working together to help an organization achieve a single goal. While at one point in time all of these layers resided on a single large system, it is quite common today for applications to be composed of multiple tiers of technology, each running on one or more systems. Technical support is challenging enough when these layers of software are running on physical systems. When they are distributed, support becomes an exciting exercise that the great Sherlock Holmes would most certainly enjoy. As organiztions host the tiers of a complex application system on virtual systems, they are often surprised to find that their technical support challenge increases dramatically . Suppliers of application development tools or applications often haven't tested their software in exactly the same environment that is in use in an organization. This also means that the suppliers haven’t trained their support engineers for this environment. So the support engineers often find dealing with virtual environments impossible.
When an organization is facing an outage, the last thing they want to hear is “We don’t support that environment. Please duplicate the issue you’re seeing in one of our supported configurations and we’ll do our best to help.” When the staff's paychecks are being held up because the payroll system crashed, this response is going to seem unacceptable to say the least.
How did we get here?Technology suppliers do their best to offer their products in the most commonly seen configurations. They follow the well known “80/20” rule in this area. That is they select and test the most commonly seen configurations (somewhere near 20% of all possible configurations) knowing that this approach will serve the needs of roughly 80% of their targeted customers.
When one considers what’s found in most datacenters that means that these suppliers must offer support of their product running in a complex mix of hardware and software environments. This typically includes several different hardware configurations and operating systems: Mainframes, Windows, several UNIX implementations and several Linux distributions.
Has your organization run into this? If so, what did you do to support your IT environment.