Power-aware applications for scientific cluster and distributed computing

TL;DR

This paper discusses strategies for reducing power consumption in scientific distributed computing systems like the WLCG, emphasizing power-aware software and scheduling to optimize energy use and costs.

Contribution

It introduces the concept of power-aware applications and scheduling in distributed computing, highlighting potential for energy savings in large-scale scientific systems.

Findings

Power-aware software can reduce energy use in distributed systems.

Scheduling optimizations contribute to lower power consumption.

Case studies include FNAL and Princeton HPC centers.

Abstract

The aggregate power use of computing hardware is an important cost factor in scientific cluster and distributed computing systems. The Worldwide LHC Computing Grid (WLCG) is a major example of such a distributed computing system, used primarily for high throughput computing (HTC) applications. It has a computing capacity and power consumption rivaling that of the largest supercomputers. The computing capacity required from this system is also expected to grow over the next decade. Optimizing the power utilization and cost of such systems is thus of great interest. A number of trends currently underway will provide new opportunities for power-aware optimizations. We discuss how power-aware software applications and scheduling might be used to reduce power consumption, both as autonomous entities and as part of a (globally) distributed system. As concrete examples of computing centers we provide information on the large HEP-focused Tier-1 at FNAL, and the Tigress High Performance Computing Center at Princeton University, which provides HPC resources in a university context.