RFS: Efficient and flexible remote file access for MPI-IO

Jonghyun Lee*, Xiaosong Ma, Robert Ross, Rajeev Thakur, Marianne Winslett

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

23 Citations (Scopus)

Abstract

Scientific applications often need to access remote file systems. Because of slow networks and large data size, however, remote I/O can become an even more serious performance bottleneck than local I/O performance. In this work, we present RFS, a high-performance remote I/O facility for ROMIO, which is a well-known MPI-IO implementation. Our simple, portable, and flexible design eliminates the shortcomings of previous remote I/O efforts. In particular, RFS improves the remote I/O performance by adopting active buffering with threads (ABT), which hides I/O cost by aggressively buffering the output data using available memory and performing background I/O using threads while computation is taking place. Our experimental results show that RFS with ABT can significantly reduce the remote I/O visible cost, achieving up to 92% of the theoretical peak throughput. The computation slowdown caused by concurrent I/O activities was 0.2-6.2%, which is dwarfed by the overall performance improvement in application turnaround time.

Original languageEnglish
Title of host publication2004 IEEE International Conference on Cluster Computing, ICCC 2004
Pages71-81
Number of pages11
DOIs
Publication statusPublished - 2004
Externally publishedYes
Event2004 IEEE International Conference on Cluster Computing, ICCC 2004 - San Diego, CA, United States
Duration: 20 Sept 200423 Sept 2004

Publication series

NameProceedings - IEEE International Conference on Cluster Computing, ICCC
ISSN (Print)1552-5244

Conference

Conference2004 IEEE International Conference on Cluster Computing, ICCC 2004
Country/TerritoryUnited States
CitySan Diego, CA
Period20/09/0423/09/04

Fingerprint

Dive into the research topics of 'RFS: Efficient and flexible remote file access for MPI-IO'. Together they form a unique fingerprint.

Cite this