Click here for full text:
Energy Flow in the Information Technology Stack: Coefficient of Performance of the Ensemble and its Impact on the Total Cost of Ownership
Patel, Chandrakant D.; Sharma, Ratnesh K.; Bash, Cullen E.; Beitelmal, Monem
Keyword(s): chip; system; data center; data center management; power; cooling; smart data center; sustainability; energy efficiency; data center metrics; TCO
Abstract: The industry is in the midst of a transformation to lower the cost of ownership through consolidation and better utilization of critical data center resources. Successful consolidation necessitates increasing utilization of capital intensive "always-on" data center infrastructure, and reducing recurring cost of power. A need exists, therefore for an end to end physical model that can be used to design and manage dense data centers and determine the cost of operating a data center. The chip core to the cooling tower model must capture the power levels and thermo-fluids behavior of chips, systems, aggregation of systems in racks, rows of racks, room flow distribution, air conditioning equipment, hydronics, vapor compression systems, pumps and cooling towers or heat exchangers. As a first step in data center consolidation, the ensemble model must be able to characterize a given data center and its level of capacity utilization, controllability, and room for expansion. Secondly, the continuous operation of the data center management system demands that the ensemble model be programmable to create new "set points" for power and cooling based on current customer cost and performance needs. The overall data center management system, when bundled as a product, must result in a simple payback of 1 year by increasing data center utilization to 80% of rated capacity and through savings in recurring cost of power. Therefore, economic ramifications drive a business need for the creation of an information technology management tool that can maximize the utilization of critical data center resources and minimize the power consumption. The creation of such an end to end management system that can sense and control a complex heat transfer stack requires a thermodynamics based evaluation model. Earlier work has outlined the foundation for creation of a "smart" data center through use of flexible cooling resources and a distributed sensing system that can provision the cooling resources based on the need. This paper shows a common thermodynamic platform which serves as an evaluation and basis for a policy based control engine for such a "smart" data center with much broader reach - from chip core to the cooling tower.
Back to Index