Understanding Application Workloads Is the Key to Cloud Computing

Time again, when the discussion turns to whether cloud computing is a worthwhile endeavor, the conversation usually ends with it all depending on the type of application workload involved.

Unfortunately, that’s not much in the way of practical help, so the folks at CiRBA have come up with the Data Center Intelligence Tool, an analytics application that allows IT organizations to analyze the nature of any application workload and the systems required to support it.

According to CiRBA CTO Andrew Hillier, the challenge that most IT organizations currently confront is that no one can tell them which applications they have running today that might be able to withstand the rigors of running on shared infrastructure. The simple fact of the matter is that most of these applications were not designed to run in a multi-tenant environment. So while cloud computing sounds attractive, it’s really only optimal for new applications that were designed with cloud computing in mind.

That doesn’t mean that all legacy applications can’t run in the cloud; it just means that IT organizations need to really understand the dynamics of those applications before moving them to the cloud. Cloud computing, especially the public kind, introduces a lot of latency issues that need to be considered. This is particularly true when any particular application workload is part of an extended business process.

Hillier says one of the primary reasons that IT is a mess these days is because the open systems revolution resulted in every application running on its own server, regardless of what the actual utilization rate of that server actually was. Cloud computing is part of an overall effort to not only make IT more flexible, but also to clean up inefficient IT infrastructure that was allowed to irrationally flourish during better economic times.

Ideally, Hillier says IT organizations will now be able to focus more on building more manageable IT environments that are optimized for average workloads. External cloud computing systems can then be called on to help out with peak application workloads. The thing to remember, says Hillier, is that not every application workload is going to peak at the same time. So rather than over provision in the age of the cloud, IT organizations should develop a more holistic view of their application workload requirements in order to get the most out of their internal private cloud infrastructure before deciding to call on an outside provider of cloud computing services for any augmentation they might need.