This audio was created using Microsoft Azure Speech Services
One of the things you’d notice if you spent as much time with DCIM users as me, is that people and companies have different motivations for installing the software and different measures for success. Sometimes a financial return on investment is targeted, sometimes ROI is viewed as an operational or productivity gain. Or it could be something as simple as added convenience – saving the time and cost associated with having somebody visit the facility and see what space is available in a rack. A prime case of financial ROI was brought to my attention at Cardiff University in the UK.
One of university’s data centers houses both general IT services as well as two supercomputer clusters: one for Advanced Research Computing at Cardiff (ARCCA) and one for High Performance Computing Wales. ARCCA was originally established to help the University maintain and build upon its position as a global centre for research. HPC Wales supports major research projects in collaboration with small and medium-sized enterprises.
Hugh Beedie, CTO at Cardiff University, told me that when they had replaced a 120kW chiller with a 300kW unit to upgrade the cooling capacity of the data center, it had somehow caused an operational imbalance in the system. The operations teams noted an increase in PUE, but could not surmise the cause. A further complication was that they did not have insight into how each element of the cooling system was performing to pinpoint the reasons for the decline in efficiency.
The prospect of new and valuable research projects coming to the University created the demand for an additional investment in cooling infrastructure to meet the capacity and resiliency requirements of a growing compute load. When making the case for a further cooling system upgrade, Hugh Beedie demonstrated that improved power efficiency, as evidenced by an improved PUE, would result in sufficient energy savings to offset the additional capital investment.
However, essential to proving the business case would be an improvement to the monitoring and analysis of all elements of the infrastructure. “The data center had an estimated annual PUE between 1.7 and 1.8 at that time but they weren’t precise numbers and they certainly weren’t being generated in real time,” said Hugh Beedie. He calculated that reducing the PUE from 1.7 to 1.4 would see the cost of the cooling upgrade pay for itself easily over the working lifetime of any new equipment.
As part of the power and cooling upgrade the University deployed Schneider Electric Data Center Operation: Energy Efficiency module as an additional component to the previously installed StruxureWare for Data Centers DCIM solution. Working with data inputs from extensive instrumentation the new software module provided a much more comprehensive picture of power and cooling consumption.
In fact, it allowed them to get deeper and more granular insights into energy usage, not just at overall site level but also at subsystem level. Critically, it did so in real time enabling them to monitor the effects on data center energy consumption of changing fan speeds or CPU utilisation, raising the temperature of the chilled water supply, or even the impact of external ambient temperatures.
“The additional information allowed me to see the real-time effect of the weather as the day progresses. For example, at the start of a summer’s day, our PUE figure is typically about 1.20, But by the end of a hot afternoon it’s up to about 1.25 – 1.27 and before dropping down again as the evening cools. This degree of detail from the DCIM enables me to be confident that we are adopting an appropriate operating regime for the cooling system.”
Dr. Beedie has long contended that “going green saves money,” so he’s pleased that the software solution is helping to verify this assertion. “The additional investment for the Energy Efficiency module enabled me to prove firstly that the business case for the cooling upgrade was vindicated, and secondly, allowed me to fine tune the services operation to achieve the best PUE possible for the mix of climate, HPC server demand and services efficiency.”
Management software available today can help data center professionals to maximize power, cooling and space capacity, control facility energy costs and utilize IT assets more effectively.
