LogfP - a model for small messages in InfiniBand

Accurate models of parallel computation are often crucial to optimize parallel algorithms for their running time. In general the easier the model's use and the smaller the number of parameters and interdependencies among them, the more inaccuracies are introduced by simplification. On the other...

Full description

Saved in:
Bibliographic Details
Published inProceedings 20th IEEE International Parallel & Distributed Processing Symposium p. 6 pp.
Main Authors Hoefler, T., Mehlan, T., Mietke, F., Rehm, W.
Format Conference Proceeding
LanguageEnglish
Published IEEE 2006
Subjects
Algorithm design and analysis
Bandwidth
Computer architecture
Computer networks
Concurrent computing
Delay
Hardware
Parallel algorithms
Parallel processing
Programming profession
Online AccessGet full text

Cover

More Information
Summary:Accurate models of parallel computation are often crucial to optimize parallel algorithms for their running time. In general the easier the model's use and the smaller the number of parameters and interdependencies among them, the more inaccuracies are introduced by simplification. On the other hand a too complex model is unusable. We show that it is possible to derive a relatively accurate and easy model for small message performance over the InfiniBand network. This model allows the developer to gain knowledge about the inherent parallelism of a specific InfiniBand hardware and encourages him to use this parallelism efficiently. Several well known models hide this feature and some of them even penalize the use of parallelism because the model designers were not aware of new emerging architectures like InfiniBand
ISBN:1424400546
9781424400546
ISSN:1530-2075
DOI:10.1109/IPDPS.2006.1639624