Using reinforcement learning to select a DS processing unit
Abstract:
A computer readable storage medium includes memory sections that store operational instructions, the when executed by one or more computing devices of a dispersed storage network (DSN), cause the one or more computing devices to perform the following for a data access request. The computing device(s) access a plurality of estimated efficiency models of a plurality of dispersed storage (DS) processing units of the DSN. The computing device(s) select one of the DS processing units based on the plurality of estimated efficiency models, a type of request, and a randomizing factor. The computing device(s) send the data access request to the selected DS processing unit for execution. The computing device(s) determine an actual processing efficiency of the processing of the data access request by the selected DS processing unit and update the estimated efficiency model of the selected DS processing module based on the actual processing efficiency.
Public/Granted literature
Information query
Patent Agency Ranking
0/0