The great paradox of cloud computing is that the primary reason most organizations embrace it is cost, and the number one issue they need to address after adopting cloud services is also cost.
A new report from 451 Research drives the point home. Recent 451 Research surveys find that 39 percent cite cost savings as the primary reason for shifting workloads to the cloud. But over half the IT organizations (53 percent) also note that their biggest pain point after adopting cloud services is containing those costs.
The reason this Jevons paradox exists comes down to exuberance, some of which is of the irrational kind. Because it’s simpler and faster to build and deploy workloads on a public cloud a lot of enthusiasm gets generated among developers and line of business (LOB) executives. Projects that never would have made it off the application development backlog list a few years ago are now getting green lighted. Just under a third of the respondents (32 percent) to 451 Research cited agility as a major benefit of moving to the cloud. Before too long the IT organization starts observing that there are more virtual machines spinning up than ever.
The problem is not all those virtual machines consistently spin down when not being employed. Left untended those virtual machines still consume resources that show up in a monthly bill. Between the increased number of workloads being deployed and what amounts to poor IT housekeeping some organizations by the end of a year can find themselves spending more on IT than when all their application workloads ran on premises. The limited availability of those on-premises resources acted a governor on the number of resources being consumed, while internal IT teams most of the time made sure those resources were not being wasted.
Moving towards cloud optimization
Because cost is a major pain point these days it’s not surprising to see so much focus on what is referred to a cloud optimization, which consist of applications that essentially makes use of monitoring tools to alert organizations when consumption of cloud resources is sub-optimal. Mostly that consists of identifying virtual machines running idle, switching workloads to lower cost virtual machine instances that are made available on a spot market, or moving long-running workloads that might be more cost effectively deployed in an on-premises environment. Public clouds are a great place to launch an application to see if it will gain traction. But once an application is shown to viable it may not make economic sense to leave it running on a public cloud that charges by the hour for the privilege.
Most internal IT organizations are not going to be especially good at cloud arbitrage, so providing a cloud optimization service creates a unique opportunity for managed service providers (MSPs). Once they begin to provide that service it’s usually not too long after that an MSP finds themselves managing more workloads on behalf of the organization they are advising. Much like a physician, a good MSP needs to determine where the source of the customer pains lie. Only then can an effective treatment plan be prescribed.
Photo: ESB Professional / Shutterstock.