TY - GEN
T1 - Starchart
T2 - 22nd International Conference on Parallel Architectures and Compilation Techniques, PACT 2013
AU - Jia, Wenhao
AU - Shaw, Kelly A.
AU - Martonosi, Margaret Rose
PY - 2013
Y1 - 2013
N2 - Graphics processing units (GPUs) are in increasingly wide use, but significant hurdles lie in selecting the appropriate algorithms, runtime parameter settings, and hardware configurations to achieve power and performance goals with them. Exploring hardware and software choices requires time-consuming simulations or extensive real-system measurements. While some auto-tuning support has been proposed, it is often narrow in scope and heuristic in operation. This paper proposes and evaluates a statistical analysis technique, Starchart, that partitions the GPU hardware/software tuning space by automatically discerning important inflection points in design parameter values. Unlike prior methods, Starchart can identify the best parameter choices within different regions of the space. Our tool is efficient - evaluating at most 0.3% of the tuning space, and often much less - and is robust enough to analyze highly variable real-system measurements, not just simulation. In one case study, we use it to automatically find platform-specific parameter settings that are 6.3× faster (for AMD) and 1.3× faster (for NVIDIA) than a single general setting. We also show how power-optimized parameter settings can save 47W (26% of total GPU power) with little performance loss. Overall, Starchart can serve as a foundation for a range of GPU compiler optimizations, auto-tuners, and programmer tools. Furthermore, because Starchart does not rely on specific GPU features, we expect it to be useful for broader CPU/GPU studies as well.
AB - Graphics processing units (GPUs) are in increasingly wide use, but significant hurdles lie in selecting the appropriate algorithms, runtime parameter settings, and hardware configurations to achieve power and performance goals with them. Exploring hardware and software choices requires time-consuming simulations or extensive real-system measurements. While some auto-tuning support has been proposed, it is often narrow in scope and heuristic in operation. This paper proposes and evaluates a statistical analysis technique, Starchart, that partitions the GPU hardware/software tuning space by automatically discerning important inflection points in design parameter values. Unlike prior methods, Starchart can identify the best parameter choices within different regions of the space. Our tool is efficient - evaluating at most 0.3% of the tuning space, and often much less - and is robust enough to analyze highly variable real-system measurements, not just simulation. In one case study, we use it to automatically find platform-specific parameter settings that are 6.3× faster (for AMD) and 1.3× faster (for NVIDIA) than a single general setting. We also show how power-optimized parameter settings can save 47W (26% of total GPU power) with little performance loss. Overall, Starchart can serve as a foundation for a range of GPU compiler optimizations, auto-tuners, and programmer tools. Furthermore, because Starchart does not rely on specific GPU features, we expect it to be useful for broader CPU/GPU studies as well.
KW - GPU
KW - auto-tuning
KW - decision tree
KW - design space exploration
KW - regression tree
UR - http://www.scopus.com/inward/record.url?scp=84887413764&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84887413764&partnerID=8YFLogxK
U2 - 10.1109/PACT.2013.6618822
DO - 10.1109/PACT.2013.6618822
M3 - Conference contribution
AN - SCOPUS:84887413764
SN - 9781479910212
T3 - Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT
SP - 257
EP - 267
BT - PACT 2013 - Proceedings of the 22nd International Conference on Parallel Architectures and Compilation Techniques
Y2 - 7 September 2013 through 11 September 2013
ER -