In the dystopian 1973 sci-fi thriller Soylent Green, detective Frank Thorn discovers one of the primary food rations of the world in 2020- Soylent Green- is made of processed human remains. “Soylent green is people” is a climactic and most famous quote from the movie.
The title of this article is not meant to be quite as macabre as the meaning of the original. It’s simply that there is more to a Cloud’s functioning than the underlying technology. The tech that underlies the Cloud is incredibly interesting, increasingly powerful and changing at a rate that generates continuous hyperbolic press releases that clog our Twitter, Facebook and email feeds. So of course, the underlying technology used to build and use the Cloud is foundational.
Yet in a world full of so much very shiny tech to be chased, we’ve observed Cloud failure that cannot (or should not) be attributed to a technology fail. These cases do exist. If you look at why one company is successful at extracting ROI from a Cloud and a similar company either fails or suffers losses, it is often not that the software failed, but rather the peoples’ interaction with that Cloud. Let’s consider some of the users that may make or break a Cloud:
Management
“Hey Ed, XYZ Corp. just got one of those Cloud thingies. We’d better get one too.” The Cloud is a shiny, hot topic and misunderstood by many in the executive class. These misunderstandings have caused failure when the wrong tool is being applied to the wrong problem or even deployed when there was no problem to solve in the first place. Understanding the use case for Cloud and its cost-benefit in solving a particular business problem are often overlooked and can cause management-generated Cloud failure.
Operations
We have a soft spot for Operations, but yes, Ops failure can also be the cause of Cloud failure to deliver sufficient ROI. Because of the cutting edge nature of many Cloud enabling technologies, there is a recognized skills gap where business are simply having a hard time finding the people with the skills to run their system efficiently. Not having enough staff to run a cloud system will slow progress at best and may impact system efficiency and success. Businesses need to recognize the importance of investing in the education of the operators and administrators that will run the system and that investing in educating existing resources is likely more effective than waiting- and waiting- and waiting- until you find someone to fill a Cloud resource gap.
Users
Yes, even (often?) the users can cause Cloud fail. A primary issue is users not being aware of or understanding the resource that is available to them in the first place and going to alternative solutions on their own. If you have a Private Cloud and your users are still going outside to use Public Cloud resources, you have a problem. The accounting will show the Private Cloud as being a poor ROI, when in fact, this may well not be the case if the external users were brought back in-house.
Another challenge is users simply not understanding the Cloud Way of using resources. If users treat VMs, IP addresses, and storage systems in a traditional manner of “it is mine and I will keep it,” then Cloud resource pools will quickly be consumed and potentially lie idle. The Cloud bloat that this causes will impact ROI. It can even be detrimental to a company’s bottom line as continued investment in Cloud infrastructure does not translate into maximal benefit.
All three categories
The final people fail is a communication and education issue. It is easy – and I’ve seen this over and over- to blame either the technology or the other user groups for a Cloud’s failings. It is important that all user groups understand the use case for a Cloud system, the capabilities of the system, and principles of good Cloud citizenship. Cross-stakeholder understanding will help all involved to more efficiently use available resources and further recognize when there is a divergence between intended use case and system function.