SC13 Denver, CO

The International Conference for High Performance Computing, Networking, Storage and Analysis

Mitigating System Noise With Simultaneous Multi-Threading.

Authors: Eli Rosenthal (Brown University), Edgar A. Leon (Lawrence Livermore National Laboratory), Adam T. Moody (Lawrence Livermore National Laboratory)

Abstract: System noise on commodity clusters significantly impacts the scalability of scientific applications. As the number of nodes in a cluster grows, this impact outweighs the benefits of the increased system capability. In this work, we propose using Simultaneous Multi-Threading (SMT) to mitigate system noise, and we demonstrate its effectiveness using Fixed Work Quantum (FWQ) micro-benchmarks. This technique relies on using the floating-point and integer units of a core (shared SMT resources) for system processing when an application is stalled waiting for an out-of-core operation. Unlike core-specialization, which dedicates a core for system processing, our approach allows an application to use all cores of a node and proves significantly more effective in reducing system noise.

Poster: pdf
Two-page extended abstract: pdf

Poster Index