It has long been suggested that the move to the cloud would increase efficiency for organizations. Since lots of IT assets (be they software or physical hardware) see little utilization, the theory goes that putting IT in the cloud ties usage closely to available resources.
A new study from Anthesis Group and a Stanford University Researcher suggests that in the hardware space, in particular, server utilization is worse than we thought.
Gartner has long talked about the "80% rule": that 80 percent of IT budgets get spent simply "keeping the lights on", this survey seeks to wrap some clarity around that. According to McKinsey and Company, typical servers in business and enterprise data centers deliver between five and 15 percent of their maximum computing output on average over the course of the year.
The research was conducted by Jonathan Koomey, Research Fellow at Stanford University, using data from TSO Logic. If Koomey's name rings a bell with cloud watchers, it is probably because he did some interesting work creating a TCO calculator that I covered a few years ago.
Koomey is back and in this study he was trying to get a handle on how much data center capacity sits idle. The core findings of the study are based on a sample of anonymized data and revealed that 30 percent of the physical servers were “comatose.” In this instance, comatose servers are those that have not delivered information or computing services in six months or more. To anyone in the cloud world, a server sat six months doing nothing is well beyond comatose, but I digress.
The survey findings imply that there are about 10 million comatose servers worldwide - including standalone servers and host servers in virtual environments. The findings support previous research performed by the Uptime Institute, which also found that around 30 percent of servers are unused. The 10 million estimated comatose servers translates into at least $30 billion in data center capital sitting idle globally (assuming an average server cost of $3,000, while ignoring infrastructure capital costs as well as operating costs).
Of course the cloud propoenents would hold up this survey as justification for moving everything to the cloud. But... not so fast.
While an idle server sitting in a data center is undoubtedly a waste of resource, it is a waste that is fairly evident. Not so much in the cloud where individuals, even low-level staff toting a personal credit card, can spin up servers with any of the cloud vendors. And once they do, those servers are sitting there, and being paid for, in perpetuity or at least until someone shuts them down.
So rather than research that justifies a move from phsycial to cloud-based infrastructure, the report rather highlights the very real need for clarity in terms of infrastructure spend. And this is, perhaps, one area where the cloud can drive a benefit. Given that the setup and management of cloud infrastructure is performed via APIs, it is conceptually easier to create clear visibility for cloud infrasturcutre spend than for phsyical hardware.
In the physical world, clarity requires manual record keeping, especially so for all those servers that are sitting dormant and hence aren't visible to management systems. In the cloud world, however, a management control panel, or a cloud spend management solution, will have access to these metrics.
Commenting on his research, Koomey points out that:
In the twenty first century, every company is an IT company, yet far too little attention is given to IT inefficiencies, and to the need for widespread changes in how IT resources are built, provisioned, and managed. Removing idle servers would result in gigawatt-scale reductions in global IT load, the displaced power use from which could then support new IT loads that actually deliver business value. That’s a result that everyone should cheer.
Clearly this is wasteful, both for the orgaizations involved, but for the environment as well. In exposing these statistics, Koomey gives us all a chance to ponder our IT infrastructure, and whether it's really delivering value.
