In-Depth
Cloud Computing in the Real World: 3 Ways to Avoid Cloud Project Failure
How to overcome the three most common impediments to cloud project success.
By David S. Linthicum
Cloud computing is all the rage within the enterprise. Enterprises are migrating data, systems, and infrastructure to cloud computing services. They are also building private clouds internally. In this torrent of activity, there are cloud projects that will do whatever it takes to succeed, and thus they make common mistakes that will cause them to fail.
Let's try to stop the insanity. Here are three common reasons why cloud computing projects fail, and how to avoid them.
Reason #1: No complete understanding of the requirements up front
Remember the old cartoon? "You guys start coding and I'll go see what they want." That still causes a few chuckles, but it seems that looking to interact your way to success has returned with a vengeance in the world of cloud computing.
To prevent project failure, you must have a complete and detailed understanding of all your requirements before you pick the technology -- or even the approach. This means logical and physical architecture, interface designs, database schemas, provisioning mechanisms, etc. From here you can allocate and plan for the resources you'll require. You just need to stick to the plan and things will be fine.
Reason #2: Jumping too quickly into the technology
Techies love technology, but not necessarily when it comes to designing and planning. They have a tendency to begin the project by evaluating and selecting the technology, then hope the requirements will fit into their technology selection. Although you can get lucky, chances are you'll pick a technology or service provider that is less than optimal or, worse, does not work.
To succeed, push the technology selection decisions down the road until you completely understand your requirements (see my previous suggestion). By doing so, you'll have a good foundation of understanding before you check out the various technologies on the market.
Reason #3: Lacking Talent.
The ability for a cloud computing project to succeed is directly related to the experience and talent of the humans on the project. Lacking the skills you need, the risk of failure goes up drastically. Although this seems obvious, those running projects are indeed having trouble finding the right talent they need to both make the right decisions and drive to the finish line. The reality is that the better cloud computing minds out there are scooped up quickly these days, and thus competition for the right people is fierce.
The best way to avoid this problem is to make sure you get the talent you need on the project, either by leveraging outside consultants or mentors, or by providing the right amount of training. Just a bit more spending here will keep your project out of a world of hurt.
A Final Word
In many respects, there is no way to completely eliminate risk from cloud projects. However, with a few simple concepts, you can remove most of the issues that will quickly kill the project. I hope you heed these warnings before you embark on the road to cloud computing.
David S. Linthicum is the founder and the CTO of Blue Mountain Labs and an internationally recognized industry expert and thought leader. He is the author or co-author of 13 books on computing, including Enterprise Application Integration (Addison Wesley). You can contact the author at david@bluemountainlabs.com.