Predicting Queue Times on Space-Sharing Parallel Computers

  title={Predicting Queue Times on Space-Sharing Parallel Computers},
  author={Allen B. Downey},
We present statistical techniques for predicting the queue times experienced by jobs submitted to a space-sharing parallel machine with rstcomerst-served (FCFS) scheduling. We apply these techniques to trace data from the Intel Paragon at the San Diego Supercomputer Center and the IBM SP2 at the Cornell Theory Center. We show that it is possible to predict queue times with accuracy that is acceptable for several intended applications. The coe cient of correlation between our predicted queue… CONTINUE READING
Highly Cited
This paper has 161 citations. REVIEW CITATIONS


Publications citing this paper.
Showing 1-10 of 106 extracted citations

Predicting application run times with historical information

J. Parallel Distrib. Comput. • 2004
View 14 Excerpts
Highly Influenced

Using Historical Data to Predict Application Runtimes on Backfilling Parallel Systems

2010 18th Euromicro Conference on Parallel, Distributed and Network-based Processing • 2010
View 6 Excerpts
Highly Influenced

161 Citations

Citations per Year
Semantic Scholar estimates that this publication has 161 citations based on the available data.

See our FAQ for additional information.


Publications referenced by this paper.
Showing 1-10 of 11 references

A comparison of workload traces from two production parallel machines

Kurt Windisch, Virginia Lo, Dror Feitelson, Bill Nitzberg, Reagan Moore
In 6th Symposium on the Frontiers of Massively Parallel Computation, • 1996
View 1 Excerpt

Paragon throughput analysis

Ken Steube, Reagan Moore
In Proceedings of the Intel Supercomputer Users Group, • 1994
View 2 Excerpts

Similar Papers

Loading similar papers…