FastForward for Efficient Pipeline Parallelism

Authors:
John Giacomoni;Tipp Moseley;Manish Vachharajani
Affiliations:
University of Colorado at Boulder, USA;University of Colorado at Boulder, USA;University of Colorado at Boulder, USA
Venue:
PACT '07 Proceedings of the 16th International Conference on Parallel Architecture and Compilation Techniques
Year:
2007

Citing 0
Cited 3

Automatic parallelization of simulink applications

Proceedings of the 8th annual IEEE/ACM international symposium on Code generation and optimization
Skewed pipelining for parallel simulink simulations

Proceedings of the Conference on Design, Automation and Test in Europe
A software-based root-of-trust primitive on multicore platforms

Proceedings of the 6th ACM Symposium on Information, Computer and Communications Security

Quantified Score

Hi-index	0.00

Visualization

Abstract

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.