Identifying Performance Bottlenecks in Work-Stealing Computations

Authors:
Nathan R. Tallent;John M. Mellor-Crummey
Affiliations:
Rice University;Rice University
Venue:
Computer
Year:
2009

Citing 0
Cited 4

Utilization of map-reduce for parallelization of resource scheduling using MPI: PRS

Proceedings of the 2011 International Conference on Communication, Computing & Security
Processing data streams with hard real-time constraints on heterogeneous systems

Proceedings of the international conference on Supercomputing
On performance analysis of a multithreaded application parallelized by different programming models using intel VTune

PaCT'11 Proceedings of the 11th international conference on Parallel computing technologies
Steal Tree: low-overhead tracing of work stealing schedulers

Proceedings of the 34th ACM SIGPLAN conference on Programming language design and implementation

Quantified Score

Hi-index	4.10

Visualization

Abstract

Work stealing is an effective load-balancing strategy for multithreading, but when comp- utations based on it underperform, traditional tools can't explain why. To resolve a computation's key performance obstacles, tools must pinpoint and quantify parallel idleness and overhead.