US 12,236,267 B2
Method and system for performing domain level scheduling of an application in a distributed multi-tiered computing environment using reinforcement learning
William Jeffery White, Plano, TX (US); and Said Tabet, Austin, TX (US)
Assigned to DELL PRODUCTS L.P., Round Rock, TX (US)
Filed by Dell Products L.P., Round Rock, TX (US)
Filed on Apr. 15, 2022, as Appl. No. 17/722,042.
Prior Publication US 2023/0333884 A1, Oct. 19, 2023
Int. Cl. G06F 9/44 (2018.01); G06F 9/48 (2006.01)
CPC G06F 9/4881 (2013.01) 17 Claims
OG exemplary drawing
 
1. A method for performing domain level scheduling in a distributed multi-tiered computing (DMC) environment, comprising:
decomposing, by a local controller associated with a DMC domain, a service dependency graph associated with a scheduling job,
wherein the service dependency graph is received from a global controller in a scheduling package that is generated for the local controller to perform the domain level scheduling in the DMC domain,
wherein the scheduling package specifies workload resource information comprised in a manifest associated with tasks, wherein the manifest is provided by a user,
wherein the workload resource information specifies a minimum clock frequency and a critical path clock frequency associated with the tasks;
assigning normalized compute units and normalized network units to the tasks specified in the service dependency graph;
generating a Q-table using the service dependency graph and reinforcement Q-learning;
calculating a critical path and a max learned path using the Q-table and the service dependency graph;
calculating an earliest start time and a latest start time for each task using the service dependency graph and the max learned path to obtain a plurality of earliest start time and latest start time pairs for each task; and
generating scheduling assignments using the plurality of earliest start time and latest start time pairs for each task.