David Schirmacher is going to present the 7x24 Exchange keynote on Tuesday, Oct 23 2018. One of the things I am doing for a change is reaching out to a few of speakers at 7x24 Exchange and chatting before hand about what they plan to present. My first post is with David Schirmacher.
David I chatted a bit. We actually talked a lot. We chatted for over 2 hours, and we are going to chat again. So how do I summarize 2 hours of discussion into a blog post?
The most recognized data center metric is PUE. Why don’t we have more metrics? There have been attempts by some industry organizations. Even though DCIM systems have been built by a wide range of companies bragging about customers, why don’t they have amazing insights backed by petabytes of data into new data center performance indicators that are the new KPI every says they are using?
PUE was invented by Chris Malone (currently at Google) and Christian Belady (currently at Microsoft) when they were both at HP. Here is a nice presentation on the history by Henry Coles from Mar 19, 2014.
If you want the original paper you can get it on the ASME site.
“Metrics and an Infrastructure Model to Evaluate Data Center Efficiency
Christian L. Belady and Christopher G. Malone
This work describes two data center efficiency metrics: Power Usage Effectiveness (PUE) and Compute Power Efficiency (CPE). PUE characterizes the fraction of the total data center power used for IT work. CPE characterizes the overall data center efficiency, considering IT equipment utilization as well as how power is used in the data center. The PUE results from three data center studies are presented here. The data suggests that a carefully designed and managed data center has a PUE of 2.0. More studies are required to determine the range of values for the typical data center. A data center infrastructure and energy cost model is presented to compare hardware costs to infrastructure and energy costs. The impact of PUE on these costs is examined to illustrate the impact of data center efficiency on the total cost of operating a data center.”
I know David Schirmacher is passionate about metrics and he has regularly been reviewing his keynote presentation. What occurred to me is what few know is David‘s background. David worked at JLL and Compass Management running an operations team which of course needed metrics. Next David spent 12 years at Goldmans Sachs as Global Head of Engineering and Critical Systems. During that time there David realized he needed software to support the use of metrics. David then joined FieldView Solutions to build a system for metrics. Digital Realty recruited David to be SVP of Operations, and he realized quickly he needed a system for metrics, and he built another system that would work for Digital’s need.
In Summary, David has spent over 30 years in data center jobs where he developed his own metrics for operations, and he built two different systems for metrics. How many other people do you know have that experience and perspective? Not many. Which is why when we chat our conversations all too often turn into 2 hour discussions. The nice thing is David has been working on his keynote to make sure he gets the ideas clearly communicated with enough time for questions in the 9:00-10:00a time slot.
Metrics come from passionate people like Chris Malone and Christian Belady. David Schirmacher is another one of those people.
“Keynote:
Unleashing the Metrics of Cost and Performance
David Schirmacher (bio)
President, Reset Advisors, LLC and
President, 7×24 Exchange International
Many billions of dollars a year are spent in the design, construction and operation of mission critical data centers. For an industry that supports the core of the global business economy and impacts the lives of virtually everyone on the planet, it is hard to believe that so little of what we do is driven by a common set of measurable metrics. The introduction of the PUE metric in 2006 was a major leap in beginning to define performance but it only addresses one slice of a very large pie. Numerous metrics have followed but none have achieved the same almost universal acceptance. In this presentation, David will discuss some of the historical challenges to metrics adoption and will take a deeper dive into the measurement of cost and performance and the tools that leading edge firms use to validate their performance.”