Title
Benchmarking the effects of operating system interference on extreme-scale parallel machines
Abstract
We investigate operating system noise, which we identify as one of the main reasons for a lack of synchronicity in parallel applications. Using a microbenchmark, we measure the noise on several contemporary platforms and find that, even with a general-purpose operating system, noise can be limited if certain precautions are taken. We then inject artificially generated noise into a massively parallel system and measure its influence on the performance of collective operations. Our experiments indicate that on extreme-scale platforms, the performance is correlated with the largest interruption to the application, even if the probability of such an interruption on a single process is extremely small. We demonstrate that synchronizing the noise can significantly reduce its negative influence.
Year
DOI
Venue
2008
https://doi.org/10.1007/s10586-007-0047-2
Cluster Computing
Keywords
Field
DocType
Microbenchmark,Noise,Petascale,Synchronicity
Extreme scale,Computer science,Synchronicity,Interference (wave propagation),Petascale computing,Operating system,Benchmarking,Distributed computing
Journal
Volume
Issue
ISSN
11
1
1386-7857
Citations 
PageRank 
References 
39
2.16
12
Authors
5
Name
Order
Citations
PageRank
Pete Beckman182248.04
Kamil Iskra264246.46
Kazutomo Yoshii324918.53
Susan Coghlan429118.09
Aroon Nataraj51017.24