The Community for Technology Leaders
Proceedings of the 22nd International Conference on Parallel Architectures and Compilation Techniques (2013)
Edinburgh, United Kingdom United Kingdom
Sept. 7, 2013 to Sept. 11, 2013
ISSN: 1089-795X
ISBN: 978-1-4799-1018-2
pp: 245-255
Janghaeng Lee , Adv. Comput. Archit. Lab., Univ. of Michigan, Ann Arbor, MI, USA
Mehrzad Samadi , Adv. Comput. Archit. Lab., Univ. of Michigan, Ann Arbor, MI, USA
Yongjun Park , Adv. Comput. Archit. Lab., Univ. of Michigan, Ann Arbor, MI, USA
Scott Mahlke , Adv. Comput. Archit. Lab., Univ. of Michigan, Ann Arbor, MI, USA
ABSTRACT
Heterogeneous computing on CPUs and GPUs has traditionally used fixed roles for each device: the GPU handles data parallel work by taking advantage of its massive number of cores while the CPU handles non data-parallel work, such as the sequential code or data transfer management. Unfortunately, this work distribution can be a poor solution as it under utilizes the CPU, has difficulty generalizing beyond the single CPU-GPU combination, and may waste a large fraction of time transferring data. Further, CPUs are performance competitive with GPUs on many workloads, thus simply partitioning work based on the fixed roles may be a poor choice. In this paper, we present the single kernel multiple devices (SKMD) system, a framework that transparently orchestrates collaborative execution of a single data-parallel kernel across multiple asymmetric CPUs and GPUs. The programmer is responsible for developing a single data-parallel kernel in OpenCL, while the system automatically partitions the workload across an arbitrary set of devices, generates kernels to execute the partial workloads, and efficiently merges the partial outputs together. The goal is performance improvement by maximally utilizing all available resources to execute the kernel. SKMD handles the difficult challenges of exposed data transfer costs and the performance variations GPUs have with respect to input size. On real hardware, SKMD achieves an average speedup of 29% on a system with one multicore CPU and two asymmetric GPUs compared to a fastest device execution strategy for a set of popular OpenCL kernels.
INDEX TERMS
Kernel, Graphics processing units, Performance evaluation, Instruction sets, Hardware, Indexes, Collaboration,speculation, multi-core, multicast, network-on-chip, router
CITATION
Janghaeng Lee, Mehrzad Samadi, Yongjun Park, Scott Mahlke, "McRouter: multicast within a router for high performance network-on-chips", Proceedings of the 22nd International Conference on Parallel Architectures and Compilation Techniques, vol. 00, no. , pp. 245-255, 2013, doi:10.1109/PACT.2013.6618821
294 ms
(Ver 3.3 (11022016))