US 12,335,335 B2
Cloud financial load optimization based on power and thermal statistics aggregation
Franck Excoffier, Katy, TX (US); Brent William Yardley, Hillsboro, OR (US); Matthew Hulsey Caughman, Houston, TX (US); Rayyan Memon, Fulshear, TX (US); and Bader Ali-Saleh, Houston, TX (US)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed on Mar. 30, 2022, as Appl. No. 17/708,487.
Prior Publication US 2023/0319135 A1, Oct. 5, 2023
Int. Cl. H04L 67/10 (2022.01); G06F 9/50 (2006.01); G06F 11/30 (2006.01); G06F 11/34 (2006.01); H04L 67/1029 (2022.01)
CPC H04L 67/1029 (2013.01) [G06F 9/5094 (2013.01); G06F 11/3058 (2013.01); G06F 11/3072 (2013.01); G06F 11/3433 (2013.01); G06F 11/3452 (2013.01); G06F 2209/508 (2013.01); G06F 2209/509 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A computer-implemented method for optimizing server-based loads between data centers, the computer-implemented method comprising:
receiving, by one or more processors, data from hardware abstract layers (HAL) associated with servers in a plurality of data centers;
filtering, by the one or more processors, data associated with a portion of the plurality of data centers based on data centers having highest power usage and thermal state conditions;
aggregating, by the one or more processors, the filtered data into groups of performance data based on association with a data center;
storing, by the one or more processors, aggregated performance data associated with the data centers in a database to persist the aggregated performance data between RESTful calls and system reboots;
creating, by the one or more processors, two-ratio statistics based on the groups of performance data;
generating, by the one or more processors, data center scores based on the two-ratio statistics;
selecting, by the one or more processors, data center sites with lowest data center scores; and
initiating, by the one or more processors, a request to transfer server-based loads from the servers associated with the filtered data to the data center sites with the lowest data center scores.