US 12,405,828 B2
Consensus-based distributed scheduler
Onur Celebioglu, Austin, TX (US); John A. Lockman, III, Granite Shoals, TX (US); and Lucas A. Wilson, Cedar Park, TX (US)
Assigned to Dell Products L.P., Round Rock, TX (US)
Filed by Dell Products L.P., Round Rock, TX (US)
Filed on Jan. 13, 2022, as Appl. No. 17/575,387.
Prior Publication US 2023/0221996 A1, Jul. 13, 2023
Int. Cl. G06F 9/50 (2006.01); G06F 9/48 (2006.01); G06F 11/34 (2006.01)
CPC G06F 9/5038 (2013.01) [G06F 9/5044 (2013.01); G06F 9/5083 (2013.01); G06F 11/3433 (2013.01); G06F 11/3466 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for servicing workload requests using a distributed system comprising data processing systems, the method comprising:
obtaining, by a first data processing system of the data processing systems making up the distributed system, a workload request for a workload;
distributing, by the first data processing system, workload request information to at least a portion of the data processing systems making up the distributed system, the workload request information being based on the workload request and the portion of the data processing systems that received the workload request information and the first data processing system make up a first group of the data processing systems making up the distributed system;
initiating, by the first data processing system, a leaderless majority consensus based scheduling process where the data processing systems making up the first group use at least the workload request information to reach a leaderless majority consensus for one data processing system from the first group to service the workload request, a final result of the leaderless majority consensus being received and verified by only the one data processing system, and an entirety of the distributed system including the first group is leaderless before and after reaching the leaderless majority consensus where none of the data processing systems is ever elected and designated by any other ones of the data processing systems as being a leader, over the other data processing systems, that is able to control one or more operations of the distributed system and of the data processing systems; and
servicing, by the first data processing system and after completion of the leaderless majority consensus based scheduling process, the workload request through the one data processing system, wherein the one data processing system schedules a future performance of the workload.