SC16 Salt Lake City, UT

DS11. Cooperative Batch Scheduling for HPC Systems

Student: Xu Yang (Illinois Institute of Technology)
Advisor: Zhiling Lan (Illinois Institute of Technology)
Abstract: The insatiable demand for computing power continues to drive the involvement of ever-growing HPC systems. The HPC systems with unprecedented scale are expected to face many challenges. Some of the most prominent challenges include energy and power, network contention and job interference, concurrency, and locality. These challenges demand great technical breakthroughs in many aspects of the HPC system's software and hardware stack. We identify three problems that we aim to solve from the perspective of batch scheduling in this research, namely, energy cost, network contention, and system fragmentation. The objective of this work is to provide new batch scheduling methodologies to solve these problems in a cooperative way.

Summary: pdf
Presentation: pdf
Poster: pdf

Doctoral Showcase Index