3 ways to screw up a cloud deployment

If you want a cloud deployment that actually works, avoid these three mistakes

screw screwdriver nuts bolts hardware

Most enterprises do not have the experience or talent to effectively take advantage of private or public cloud resources, so adoption is a trial-and-error process. Even if you use case studies from other enterprises with similar problem domains, you'll still run into issues that are unique to your enterprise, including problems with data integration, governance, and poor application design.

When enterprises deploy their first cloud projects, there are three outcomes:

  • ImplementIng something that's not a cloud (such as virtualized servers); enterprise IT calls it a cloud, then declares success
  • Implementing public or private cloud services in improper ways; enterprise IT falls on its face, then declares success
  • Implementing cloud-based applications and other resources successfully; enterprise IT quickly moves on to the next cloud projects

The patterns of success, as well as the patterns of failure, are becoming better understood. Sometimes the most important lessons come from failure. Thus, here are my top three ways to screw up a cloud deployment.

1. Failure to understand the objectives before tossing technology at problems

Many companies looking to implement the cloud typically lead with technology rather than requirements. They're unlikely to have conversations around data security and governance requirements, or core business processes that are paramount to the success of the company. Instead all the talk is about Amazon Web Services, Microsoft, Google, and others.

The attitude is "Let's play with the technology first, and figure out what we're doing with it later."

2. Failure to consider performance

Clouds perform well -- OK, they perform well when used for specific scenarios with well-designed applications. Enterprises often consider the ability for cloud to self- and autoprovision as a path to performance. But the ability to provision resources does not mean those resources will perform well.

Much of the performance gained with cloud computing comes from good application design rather than merely from the cloud's ability to add more virtual servers as needed.

3. Failure to consider security, governance, and compliance

Although you would think that any company moving to the cloud would have security on its mind front and center, the reality is it's an afterthought at many enterprises.

Security has to be systemic and engineered into the cloud solution from the start. Retrofitting security typically won't work, and where it does work it often does not work well.

By understanding ways to fail in the cloud, the paths to success become clearer. Understand your business requirements before you consider vendors. Make certain your applications are designed to take advantage of cloud efficiencies. Build security, governance, and compliance requirements into the system from the beginning of the planning stages.

Copyright © 2015 IDG Communications, Inc.