Exploring RM-Replay for Cluster Scheduling

The objective of this project is twofold: (1) to learn and explore RM-Replay, a high-fidelity tuning, optimization and exploration tool for resource management and job scheduling, and (2) to compare RM-Replay with the event-driven, trace-based scheduling simulator CQSim.

The project is conducted by two senior undergraduate students, Zhen Huang and Blake Ehrenbeck, during the Spring'19 (Jan. 14 - April 26, 2019) under the supervision of Zhiling Lan. Two doctoral students, Boyang Li and Yuping Fan, are actively participating in the project discussion and design. This webpage lists their bi-weekly reports.

Bi-weekly reports:
  • report 1 on 2/3 [Link]
  • report 2 on 2/7 [Link]
  • report 3 on 2/13 [Link]
  • report 4 on 2/28 [Link]
  • report 5 on 3/14 [Link]
  • report 6 on 3/27 [Link]
  • report 7 on 4/11 [Link]
  • final report on 5/11 [Link]