Title | ||
---|---|---|
SCN-DRL: Scheduler for large-scale Critical Notification applications based on Deep Reinforcement Learning |
Abstract | ||
---|---|---|
A multi-objective Scheduling approach for large-scale microservice Critical Notification system applications (SCN-DRL) based on Deep Reinforcement Learning is presented. This paper addresses optimization for three objectives: cloud service cost, cloud resource utilization, and system notification deadline. Recent advances in the deep reinforcement learning field inspired us to research how to build a solution that learns to manage the cloud container’s resources directly from experience. SCN-DRL is an example solution that translates the containerized microservice cloud cluster scheduling problem into a learning problem. To the best of our knowledge, this is the first work dealing with multi-optimization objectives for critical microservice notification applications using deep reinforcement learning, where the notification load is variable and depends on the results of the other previously prior processed microservice. Our initial results show that SCNDRL performs better than state-of-the-art heuristics and adapts to different workload conditions without a noticeable drop in performance. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1109/FiCloud57274.2022.00026 | 2022 9th International Conference on Future Internet of Things and Cloud (FiCloud) |
Keywords | DocType | ISBN |
deep reinforcement learning,cloud computing,microservice,container orchestration,multi-objective,deadline-aware notification system,network latency | Conference | 978-1-6654-9351-2 |
Citations | PageRank | References |
0 | 0.34 | 1 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mira Vrbaski | 1 | 2 | 1.12 |
Miodrag Bolic | 2 | 503 | 58.17 |
Shikharesh Majumdar | 3 | 0 | 0.34 |