Paper Info
Title
An Optimized Task-Based Runtime System For Resource-Constrained Parallel Accelerators
Abstract
Manycore accelerators have recently prove a promising solution for increasingly powerful and energy efficient computing systems. This raises the need for parallel programming models capable of effectively lever. aging hundreds to thousands of processors. Task-based parallelism has the potential to provide such capabilities, offering flexible support to fine-gramed and irregular parallelism. However, efficiently supporting this programming paradism on resource-constramed parallel accelerators is a challenging task. In this paper, we present an optimized implementation of the OpenMP tasking model for embedded parallel accelerators, discussing the key design solution that guarantee small memory (footprint) and minimize performance overheads. We validate our design by comparing to several state-of-the-art tasking implementations, using the most representative parallelization patterns. The experimental results confirm that our solution achieves near-ideal speedups for tasks as small as 5K cycles.
Year
Venue
Field
2016
PROCEEDINGS OF THE 2016 DESIGN, AUTOMATION & TEST IN EUROPE CONFERENCE & EXHIBITION (DATE)
Implicit parallelism,Programming paradigm,Task parallelism,Efficient energy use,Computer science,Instruction set,Parallel computing,Real-time computing,Implementation,Data parallelism,Runtime system
DocType
ISSN
Citations 
Conference
1530-1591
0
PageRank 
References 
Authors
0.34
11
3
Name
Order
Citations
PageRank
Daniele Cesarini1154.96
Andrea Marongiu233739.19
Luca Benini3131161188.49