Abstract
The present disclosure relates to distributed container orchestration and, more particularly, to a First-In-First-Out (FIFO) on-demand queuing container lifecycle orchestration system for executing heavy workloads. The disclosed system receives on-demand workload requests and queues the requests based on submission time for processing by a distributed worker pool. Each worker node is restricted to executing a single workload at a time and acquires execution locks to prevent concurrent container execution conflicts. Upon selection, a worker node dynamically creates a container to execute the workload, records execution results and operational logs, and destroys the container upon completion to release computational resources. A locking-based queuing mechanism ensures orderly execution while preventing workload interference. The system further supports dynamic scaling through registration and de-registration of worker nodes, enabling efficient utilization of limited compute resources for intermittent, resource-intensive workloads.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Goh, Chang Kang Charles and Mahapatra, Hitika, "DISTRIBUTED FIFO ON-DEMAND QUEUING CONTAINER LIFECYCLE ORCHESTRATION SYSTEM FOR HEAVY WORKLOADS", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/9252