Back to results
US 9,959,371 B2Granted
Method for efficient designing and operating cooling infrastructure in a data center
AssigneeTata Consultancy Services Limited
InventorsAmarendra K. Singh, Harshad Bhagwat, Umesh Singh, Rohan Amarnath, Anand Sivasubramaniam
Life of this patent
Estimating the term…
Abstract
A method and system is disclosed for maintaining Power Usage Effectiveness (PUE) of a new data center constant or within narrow range around efficient level during ramping up stage of the data center. The method comprises of capturing a plurality of design and operational parameters of the data center, computing an efficient design for the data center at full occupancy, and maintaining the Power Usage Effectiveness constant or within narrow range around efficient level at a current occupancy during a ramp up period of the data center.
Claims12
A method for efficient configuration of a data center at full occupancy, having an initial empty occupancy level, the method comprising:
- capturing a plurality of design parameters of a data center at an empty occupancy level;
- identifying, based on the design parameters, a plurality of partitioned designs for the data center at the empty occupancy level;
- determining a plurality of rack placements and heat load distribution amongst racks within each of the partitioned designs;
- mapping cooling infrastructure parameters associated with each of the rack placements and the heat load distribution for at least one partitioned design;
- analyzing cooling performance for the mapped cooling infrastructure parameters using a Computational Fluid Dynamics (CFD) tool;
- determining Power Usage Effectiveness (PUE) using a PUE predictor tool for each of the partitioned designs by integrating an information technology (IT) power predictor to calculate power consumed by an IT equipment upon classifying the IT equipment based on type and manufacturer, a cooling power predictor to calculate power consumed by the cooling infrastructure taking into account of design parameters and operating parameters related to the cooling infrastructure, and a power losses predictor to calculate power consumed in lighting and losses in power distribution by estimating correlation of efficiency and loading level of an equipment, and calculating a total data center power as sum of an IT power calculated from the IT power predictor, a cooling power calculated from the cooling power predictor, and power losses calculated from the power losses predictor;
- and
- configuring the data center based on a partitioned design among the partitioned designs having optimum cooling performance and lowest PUE upon determining efficient combination of the rack placement, the heat load distribution and cooling related variables.
5 more elements
A processor implemented method for dynamically maintaining Power Usage Effectiveness (PUE) of a data center constant or within a narrow range around an efficient level during a ramp up period, the method comprising:
- capturing a plurality of design and operational parameters at an occupancy of a data center at a point in time;
- determining placements of servers based on the design and operational parameters;
- mapping cooling infrastructure parameters for the placements of servers;
- analyzing cooling performance for the placements of servers using a CFD tool;
- determining PUE using a PUE predictor tool for the placements of servers by integrating an information technology (IT) power predictor to calculate power consumed by an IT equipment upon classifying the IT equipment based on type and manufacturer, a cooling power predictor to calculate power consumed by the cooling infrastructure taking into account of design parameters and operating parameters related to the cooling infrastructure, and a power losses predictor to calculate power consumed in lighting and losses in power distribution by estimating correlation of efficiency and loading level of an equipment, and calculating a total data center power as sum of an IT power calculated from the IT power predictor, a cooling power calculated from the cooling power predictor, and power losses calculated from the power losses predictor;
- and
- configuring and operating the data center based on a placement of servers among the placements having lowest PUE upon determining efficient combination of a rack placement, a heat load distribution and cooling related variables, and optimum cooling performance.
4 more elements
A system for efficient configuration of a data center at full occupancy, having an initial empty occupancy level, the system comprising:
- a memory device that stores a set of instructions;
- and
- at least one processor to execute the instructions to:
- capture a plurality of design parameters of a data center at an empty occupancy level;
- identify, using the design parameters, a plurality of partitioned designs for the data center at the empty occupancy level;
- determine a plurality of rack placements and heat load distribution amongst racks within each of the partitioned designs;
- map cooling infrastructure parameters associated with each of the rack placements and the heat load distribution for at least one partitioned design;
- analyze cooling performance for the mapped cooling infrastructure parameters using a Computational Fluid Dynamics (CFD) tool;
- determine Power Usage Effectiveness (PUE) using a PUE predictor tool for each of the partitioned designs by integrating an information technology (IT) power predictor to calculate power consumed by an IT equipment upon classifying the IT equipment based on type and manufacturer, a cooling power predictor to calculate power consumed by the cooling infrastructure taking into account of design parameters and operating parameters related to the cooling infrastructure, and a power losses predictor to calculate power consumed in lighting and losses in power distribution by estimating correlation of efficiency and loading level of an equipment, and calculating a total data center power as sum of an IT power calculated from the IT power predictor, a cooling power calculated from the cooling power predictor, and power losses calculated from the power losses predictor;
- and
- configure the data center based on a partitioned design among the partitioned designs having optimum cooling performance and lowest PUE upon determining efficient combination of the rack placement, the heat load distribution and cooling related variables.
8 more elements
Dependencies are inferred from the claim text.