Impact of Data Placement on Parallel I/O Systems

The I/O performance of several concurrent external merge jobs sharing a parallel I/O system is studied. The placement of the runs is found to have a significant impact on performance. Placements leading to serialization among jobs were identified and analyzed, and solutions discussed. Contrary to th...

Full description

Saved in:
Bibliographic Details
Published in1993 International Conference on Parallel Processing - ICPP'93 Vol. 3; pp. 276 - 279
Main Authors Sinclair, J.B., Tang, J., Varman, P.J., Iyer, B.R.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.08.1993
Subjects
Database systems
Decision support systems
Degradation
Delay
Merging
Packaging
Parallel processing
Prefetching
Sorting
Online AccessGet full text

Cover

More Information
Summary:The I/O performance of several concurrent external merge jobs sharing a parallel I/O system is studied. The placement of the runs is found to have a significant impact on performance. Placements leading to serialization among jobs were identified and analyzed, and solutions discussed. Contrary to the behavior of a single merge, increasing the buffer size in this situation may actually degrade performance.
ISBN:0849389836
9780849389832
ISSN:0190-3918
2332-5690
DOI:10.1109/ICPP.1993.100