Q&A with Chuck Tatham of CiRBA

VSM: It seems like some organizations have skipped virtualization and are moving straight to the cloud. What do you think about this approach?

CT: We liken it to the rapid adoption of cellular phone service in developing countries. People are skipping traditional landlines and going mobile. Some organizations that have been behind the curve in adopting virtualization have done the same thing.

A recent survey done by CiRBA of 100 IT executives at large, global organizations found only 17 percent have met all established targets for virtualization. The danger is that the cloud can present even more challenges in achieving certain objectives such as cost savings, so it’s not necessarily a slam dunk in terms of strategies. Both internal and external clouds offer a lot of benefits, but we recommend that decisions made on which options are best be based on a thorough examination of the workloads in the environment against the available options and objectives. This might sound logical and straightforward, but it’s actually a pretty big challenge when you consider that operational policies like service levels, regulatory constraints, etc., need to be factored in as well as pure capacity and configuration requirements, not to mention costs.

VSM: What have we learned from virtualization that can now be applied to the Cloud?

CT: Most organizations have now learned that virtualizing without a strategy and without understanding the dynamics of your workloads does not lead to a path of efficiency. Many experts believe the average utilization in virtual infrastructure is about 25 percent. That’s a far cry from the 50 to 60 percent people were targeting.

Like virtualization, the key to success in the cloud is investing in a methodical, consistent process for analyzing workloads, applying pre-defined, documented policies to decisions, and placing workloads appropriately – whether in a cloud or somewhere better suited (maybe on a VM or even a physical server). This kind of process not only enables consistency in determining where things should go, but it also applies the discipline that is required to ensure that decisions are based on all the critical criteria and that a “cloud first” policy is applied only where and when it makes sense. One of our customers has aptly called this process “The Cloud Factory.” Truly fitting considering the fact that they were processing 100,000 workloads!

VSM: What are the biggest challenges in determining what workloads to move and where to move it to?

CT: These decisions really need to be grounded in the practical realities of an organization’s workloads, policies and available infrastructure. Whether you’re looking at traditional virtualization, a catalog-based standardized virtual environment or true cloud infrastructure, no company has the resources to evaluate all these parameters manually, or using spreadsheets or capacity planning tools, to figure out which options are best. We have heard of customers taking up to 65 days to analyze a workload to determine its profile and decide what cloud instance, software stack or container it should go into. Although this organization was sophisticated with a lot of capability in this area, even they struggled with digging through the data and trying to analyze in any reasonable amount of time to come up with an answer that was good enough to be actionable.