EVENTS
Home > EVENTS > Content
Static Energy Management in Supercomputer Interconnection Networks Using Topology-Aware Partitioning

Topic of Lecture:Static Energy Management in Supercomputer Interconnection Networks Using Topology-Aware Partitioning

Time of Lecture:June 25, 2018 10:30-11:30 a.m.

Location of Lecture:Mingli Building B306

Organizer:SWPU Department of Science and Technolopgy, School of Computer Science

Lecturer:Chen Juan

Abstract:With the parallel systems being scaled-up, the static energy consumed by their interconnection networks has been increasing substantially. The key to reducing static energy in supercomputers is switching off their unused components. Routers are the major components of a supercomputer. Whether routers can be effectively switched off or not has become the key to static energy management for supercomputers. For many typical applications, the routers in a supercomputer exhibit low utilization. However, it is very difficult to switch the routers off when they are idle. By analyzing the router occupancy in time and space, we present a routing-policy guided topology partitioning methodology to solve this problem. We propose topology partitioning methods for three kinds of commonly used topologies (mesh, torus and fat-tree) equipped with the three most popular routing policies (deterministic routing, directionally adaptive routing and fully adaptive routing). Based on the above methods, we propose the key techniques required in this topology partitioning based static energy management in supercomputer interconnection networks to switch off unused routers in both time and space dimensions. Three topology-aware resource allocation algorithms have been developed to handle effectively different job-mixes running on a supercomputer. We validate the effectiveness of our methodology by using Tianhe-2 and a simulator for the aforementioned topologies and routing policies.

About the Lecturer:Dr. Chen Juan, associate professor and master mentor at School of Computer Science, National University of Defense Technology, co-chairman of ACM TURC 2018 SIGCSE Program Committee, program committee member of SIGCSE '17, SIGCSE '18, ITiCSE '17, ACM TURC (SIGCSE China) '17 - '18, ICESS '14 - '16 and HPCC '08, reviewer of IEEE TPDS, Journal of Supercomputing, Frontiers of Computer Science in China and IEEE Systems Journal, and editorial board member of Tsinghua Science and Technology (High-Performance Computation).

Fields of Research:Low-power optimization technologies for large-scale parallel computer system software, energy efficiency optimization technologies, power-aware parallel algorithms, energy optimization methods for high-performance connections, optimization and scalability research on large-scale scientific computation applications, GPU/Intel MICs-based performance optimization, GPU-based energy efficiency optimization, low-power optimization for heterogeneous CPU-GPU systems, energy modelling and prediction technologies, machine learning-based energy-efficient task scheduling methods.

Previous:Inversion of the Earth – Exploration in the Earth’s Largest Unsteady State Motions Next:Sino-German Photovoltaic Forum – Symposium on High Efficiency Silicon Solar Cells and Perovskite Tandem Technologies

close