16th International Conference on Parallel Architecture and Compilation Techniques (PACT 2007) (2007)
Sept. 15, 2007 to Sept. 19, 2007
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/PACT.2007.36
John Giacomoni , University of Colorado at Boulder, USA
Tipp Moseley , University of Colorado at Boulder, USA
Manish Vachharajani , University of Colorado at Boulder, USA
High-rate core-to-core communication is critical for efficient pipeline-parallel software architectures. This paper introduces FastForward, a software-only low-overhead high-rate queue algorithm for pipeline parallelism on multicore architectures. FastForward uses an architecturallytuned domain-specific adaptation of concurrent lock-free queues to provide low-latency and low-overhead core-to-core communication. Enqueue and dequeue times on a 2 GHz Opteron 270 based system are as low as 36 ns, up to 4x faster than Lamport?s solution.
T. Moseley, M. Vachharajani and J. Giacomoni, "FastForward for Efficient Pipeline Parallelism," 16th International Conference on Parallel Architecture and Compilation Techniques (PACT 2007)(PACT), Brasov, Romania, 2007, pp. 407.