| CPC G06F 16/122 (2019.01) [G06F 16/164 (2019.01); G06F 16/1824 (2019.01)] | 20 Claims |

|
1. A computer-implemented method (CIM) for mitigating latency associated with static policy rules in a file system used in a data storage system, the CIM comprising:
training an artificial intelligence (AI)-based policy engine, wherein the AI-based policy engine is trained using a collection of historical data pertaining to activities on the file system;
causing the trained AI-based policy engine to monitor predetermined metrics, the predetermined metrics including use of a first storage pool of the data storage system, wherein the first storage pool has relatively faster performance than a second storage pool of the data storage system;
causing the trained AI-based policy engine to dynamically predict, based on the monitoring of the predetermined metrics, users that are likely to use the data storage system;
dynamically categorizing the predicted users according to a plurality of predetermined categories including: a first category of user that does not create any files of at least a predetermined large size, a second category of user that creates at least some but less than a predetermined number of files of at least the predetermined large size, and a third category of user that creates at least the predetermined number of files of at least the predetermined large size; and
dynamically assigning files of the predicted users to predetermined storage pools of the data storage system, including the first storage pool, based on the dynamic categorizations.
|