Abstract | ||
---|---|---|
Core specialization is currently one of the most promising ways for designing power-efficient multicore chips. However, approaching the theoretical peak performance of such heterogeneous multicore architectures with specialized accelerators, is a complex issue. While substantial effort has been devoted to efficiently offloading parts of the computation, designing an execution model that unifies all computing units is the main challenge. We therefore designed the StarPU runtime system for providing portable support for heterogeneous multicore processors to high performance applications and compiler environments. StarPU provides a high-level, unified execution model which is tightly coupled to an expressive data management library. In addition to our previous results on using multicore processors alongside with graphic processors, we show that StarPU is flexible enough to efficiently exploit the heterogeneous resources in the Cell processor. We present a scalable design supporting multiple different accelerators while minimizing the overhead on the overall system. Using experiments with classical linear algebra algorithms, we show that StarPU improves programmability and provides performance portability. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1007/978-3-642-03138-0_36 | SAMOS |
Keywords | Field | DocType |
execution model,heterogeneous resource,starpu unified runtime system,performance portability,multicore processor,heterogeneous multicore processor,heterogeneous multicore architecture,theoretical peak performance,high performance application,power-efficient multicore chip,starpu runtime system,linear algebra,power efficiency,chip,multicore processors,data management | Linear algebra,Computer architecture,Computer science,Parallel computing,Compiler,Exploit,Real-time computing,Execution model,Software portability,Multi-core processor,Data management,Runtime system | Conference |
Volume | ISSN | Citations |
5657 | 0302-9743 | 7 |
PageRank | References | Authors |
1.27 | 7 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Cédric Augonnet | 1 | 649 | 28.65 |
Samuel Thibault | 2 | 716 | 35.58 |
Raymond Namyst | 3 | 1405 | 83.04 |
Maik Nijhuis | 4 | 21 | 4.19 |