When I was first introduced to Romonet at an Uptime Institute Symposium last year I thought there could be good synergy between our critical power systems analytics and their data center TCO (total cost of ownership) analytics, primarily surrounding extending battery life and replacing frequency based equipment maintenance programs with predictive or condition based maintenance programs. What I wasn’t expecting was the depth of the solutions our partnership can deliver to our data center clients.
I use the word "depth" and not "multitude" for a reason. While a combined service
offering will no doubt increase the number of services provided under one contract, Romonet and Canara services complement one another creating enhanced service offerings that neither party could deliver alone. It’s why the partnership makes so much sense.
Having spent the past 13 years operating data centers, I am intrigued by the tools and information Romonet places at their client’s fingertips that can be leveraged from data center conception to end-of-life.
The magic lies in their infrastructure modeling application. Within days, if not hours, the most complex of systems designs can be modeled within their software, giving light to what can be expected from a TCO perspective the day the site is turned on.
This can provide great value even before the site is turned on. Take site selection for example. If you want to know what the optimal design is for a given region, simply modify the model to provide a side-by-side comparison for as many design options as you like. It’s quick and easy to do. Likewise, in a similar manner, they make what-if analysis a snap.
Romonet modeling can also be applied to commissioning. If you know how a site is expected to perform, it only makes sense to compare those values to actual performance values during the commissioning phases. Costly system anomalies that may otherwise be accepted as the “norm” can be corrected day-one. The Romonet model easily breaks “expected performance” down by component allowing you to identify where the anomalies lie, making it a powerful troubleshooting aid.
So what happens after the data center is handed over to the operators and the engineers and commissioning agents have left the building? How do you know if a data center is operating efficiently on a day-to-day basis? A standard benchmark is Power Utilization Effectiveness (PUE), however, we all know that comparing one site’s PUE to another’s is comparing apples to oranges. The method I used when I managed data centers was to calculate the design PUE as a target to aim at, and then track the actual PUE hoping for progressive improvements toward that target. What I lacked was a near real-time, season-to-season, hour-by-hour benchmark to compare “actual vs. expected” performance. And, to have the means to analyze any variances down to the component level so I could flush out the root cause and correct it. That’s what Romonet can do.
Now you have your data center dialed in and you have identified a long list of improvements that you want to implement that will improve your operations and “make it run more efficiently”. You’ve calculated the ROI based on component efficiency gains, but how will that impact your TCO? Are you sure there won’t be a negative trade-off with another system within your infrastructure? A better question may be, “how are you going to present proof of your cost savings to Finance”? Don’t forget this is a year-over-year endeavor and you have many other irons in the fire. Romonet not only provides a means of presenting an accurate business case for capital investments but also provides a means to track and present their progress to the stakeholders.
As long as we are talking about Finance, I don’t want to leave out billing. As the industry moves toward a cost-based structure, data center operators will want to see cost breakdowns for clients, business units, applications, on down to infrastructure components themselves. Romonet modeling can track and report on cost contributors from the component level to the data floor, providing a true cost analysis that encompasses the entire infrastructure. Combine that with branch circuit monitoring data and accurate billing for usage and tracking margin per client, and/or business unit, becomes achievable.
What initially appeared to be “good” synergy turned out to be “great” synergy. Wish I had met them years ago.