Loading [MathJax]/extensions/MathZoom.js
Profiling and Improving I/O Performance of a Large-Scale Climate Scientific Application | IEEE Conference Publication | IEEE Xplore

Profiling and Improving I/O Performance of a Large-Scale Climate Scientific Application


Abstract:

Exascale computing systems are soon to emerge, which will pose great challenges on the huge gap between computing and I/O performance. Many large-scale scientific applica...Show More

Abstract:

Exascale computing systems are soon to emerge, which will pose great challenges on the huge gap between computing and I/O performance. Many large-scale scientific applications play an important role in our daily life. The huge amounts of data generated by such applications require highly parallel and efficient I/O management policies. In this paper, we adopt a mission-critical scientific application, GEOS-5, as a case to profile and analyze the communication and I/O issues that are preventing applications from fully utilizing the underlying parallel storage systems. Through in-detail architectural and experimental characterization, we observe that current legacy I/O schemes incur significant network communication overheads and are unable to fully parallelize the data access, thus degrading applications' I/O performance and scalability. To address these inefficiencies, we redesign its I/O framework along with a set of parallel I/O techniques to achieve high scalability and performance. Evaluation results on the NASA discover cluster show that our optimization of GEOS- 5 with ADIOS has led to significant performance improvements compared to the original GEOS-5 implementation.
Date of Conference: 30 July 2013 - 02 August 2013
Date Added to IEEE Xplore: 24 October 2013
ISBN Information:
Print ISSN: 1095-2055
Conference Location: Nassau, Bahamas

I. Introduction

Scientific applications are playing a critical role in improving our daily life. They are designed to solve pressing scientific challenges, including designing new energy-efficient sources [2] and modeling the earth system [4]. To boost the productivity of scientific applications, High-Performance Computing (HPC) community has built many supercomputers [7] to provide unprecedented computation power over the past decade. Meanwhile, computer scientists are also arduously improving parallel file systems [11], [23] and I/O techniques [19], [20] to bridge the gap between fast processors and slow storage systems. However, despite the rapid evolution of HPC infrastructures, the development of scientific applications dramatically lags behind in leveraging the capabilities of the underlying systems, especially the superior I/O performance.

References

References is not available for this document.