US 11,055,234 B2
Managing allocation and demotion of cache segments between a global queue and a plurality of local queues by training a machine learning module
Lokesh M. Gupta, Tucson, AZ (US); Kevin J. Ash, Tucson, AZ (US); Beth A. Peterson, Tucson, AZ (US); and Matthew R. Craig, Sahuarita, AZ (US)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed on May 21, 2019, as Appl. No. 16/418,914.
Prior Publication US 2020/0371959 A1, Nov. 26, 2020
Int. Cl. G06F 12/126 (2016.01); G06F 12/0895 (2016.01); G06K 9/62 (2006.01); G06N 20/00 (2019.01); G06F 12/123 (2016.01)
CPC G06F 12/126 (2013.01) [G06F 12/0895 (2013.01); G06F 12/123 (2013.01); G06K 9/6256 (2013.01); G06N 20/00 (2019.01)] 23 Claims
OG exemplary drawing
 
1. A computer program product for determining cache segments in a cache to transfer between local queues used by processing units to a global queue, wherein the computer program product comprises a non-statutory computer readable storage medium having computer readable program code embodied therein that when executed performs operations, the operations comprising:
providing a machine learning module input comprising cache segment management information, related to management of segments in the local queues by the processing units and accesses of the global queue, to transfer cache segments between the local queues and the global queue to output an optimum number parameter comprising an optimum number of segments to maintain in a local queue and a transfer number parameter comprising a number of cache segments to move between a local queue and the global queue;
adjusting the transfer number parameter for at least one of the processing units to produce an adjusted transfer number parameter;
adjusting the optimum number parameter for at least one of the processing units to produce an adjusted optimum number parameter;
retraining the machine learning module based on the cache segment management information to output the adjusted transfer number parameter and the adjusted optimum number parameter for the processing units; and
using the retrained machine learning module to output the transfer number parameter and the optimum number parameter for the processing units to use to manage transferring cache segments between the local queues and the global queue.