Thu, 12/20/2012 - 13:03 — sheldon
IBM just released a new whitepaper summarizing their experiences with transitioning their internal business environment to the cloud. One of their takeaway points is something that I've had to present to clients and friends on numerous occasions recently. It's that the "cloud" is not just like getting a better and faster computer. It has its own set of risks and rewards and may not be worth the hassle if your upside for "going cloud" is limited.
When applied to the right workloads, cloud can deliver game-changing value. When applied to the wrong workloads, cloud’s value over traditional delivery models can be diminished or lost altogether.
What kind of jobs aren't ideal for cloud deployment?
Workloads that are unusually complex may offer big potential for improvement, but they may require a high degree of customization or application redesign to conform to the cloud’s architecture. These kinds of workloads can prove too difficult, risky or costly to move to the cloud. Certainly, the risk/reward profile of any workload should be clearly understood before it is deployed in the cloud. Consider complex ERP (enterprise resource planning) applications. The mission-critical nature of these production workloads raises the risk associated with cloud deployment, in some cases, far outweighing the potential gains.
Similarly, legacy and heterogeneous applications, and workloads that involve a high amount of data transfer, may be difficult to deploy in the cloud, relative to their anticipated gain.
In other words, "if it ain't broke don't fix it." A service in the cloud is not going to carry any inherent advantages over that same service deployed in a more traditional environment if that service is already being well served for now and the forseeable future. It's all about the risk-reward ratio. Even positive change carries risk so if you're moving to the cloud for still theoretical rewards, you might be better served to sit tight for a while longer.