A ‘Cool’ Load Balancer for Parallel Applications

International Conference for High Performance Computing, Networking, Storage and Analysis (SC) 2011
Pulication Type: Paper
Download: pdf ps

Abstract

Meeting the power requirements for future Exascale computing machines is one of the major challenges. Power consumption in a high performance computing (HPC) data center comprises machine power and cooling power. Earlier work has focused on minimizing machine power consumption but not much attention has been paid to minimize the power spent on cooling. We propose a dynamic frequency and voltage scaling (DVFS) based technique ensuring all cores in a job run below a user defined temperature threshold. This control comes at the cost of execution time penalty, which is amplified in tightly coupled parallel applications because a single slow-running core can slow the entire application down. We develop a strategy that minimizes this penalty. We present results from a 160 cores cluster with a dedicated air conditioning unit. Not only can we constrain the average core temperatures, we are also able to reduce chances of hot spots.

Text Ref


						

BibTex

@INPROCEEDINGS{TempLDBSC11,
        author = "Osman Sarood and Laxmikant V.~Kal{\'e}",
        title = "A `Cool' Load Balancer for Parallel Applications",
        booktitle = "Proceedings of the 2011 ACM/IEEE conference on Supercomputing",
        month = "November",
        year = "2011",
        address = "Seattle, WA",
}