Spatter: A Benchmark Suite for Evaluating Sparse Access Patterns
Spatter: A Benchmark Suite for Evaluating Sparse Access Patterns
复制标题
Spatter：用于评估稀疏访问模式的基准套件
DOI：
--
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
Jeffrey S. Young
中科院分区：
文献类型：
--
作者：
Patrick Lavin;Jason Riedy;Richard W. Vuduc;Jeffrey S. Young
数据移动性能的最新表征已经评估了GPU和Xeon Phi等加速器使用的密集和阻塞访问的优化，但是在当前和新兴的架构中，仍然对散布和收集等稀疏访问模式尚未充分了解。我们提出了一个可调的基准套件，Spatter，该套件使用户可以在多个后端（包括Cuda，OpenCL和OpenMP）的低水平表征散点，收集和相关的稀疏访问模式。 Spatter还允许用户更改移动的块大小和数据量，以创建稀疏访问模式的更全面的图片以及在实际应用程序中发现的模型。使用Spatter，我们旨在通过评估访问的密度与现实世界有效的记忆带宽（通过流衡量）的比较，以及如何在包括GPU和X86，X86，X86，，如何比较它，以新颖的方式表征记忆系统的性能。手臂和电源CPU。我们展示了如何使用Spater来生成比较不同架构的分析图，并证明当前的GPU系统可实现高达65％的流带宽，以实现稀疏访问，并且对于几种不同的稀疏模式而言，它更有效地这样做。我们的Spater Benchmark的未来计划是使用这些结果来预测新的内存访问原语对各种体系结构的影响，开发FPGA和EMU Chick（例如EMU Chick）的新型硬件的后端，以及自动测试，以便用户可以执行自己的稀疏访问研究。
Recent characterizations of data movement performance have evaluated optimizations for dense and blocked accesses used by accelerators like GPUs and Xeon Phi, but sparse access patterns like scatter and gather are still not well understood across current and emerging architectures. We propose a tunable benchmark suite, Spatter, that allows users to characterize scatter, gather, and related sparse access patterns at a low level across multiple backends, including CUDA, OpenCL, and OpenMP. Spatter also allows users to vary the block size and amount of data that is moved to create a more comprehensive picture of sparse access patterns and to model patterns that are found in real applications. With Spatter we aim to characterize the performance of memory systems in a novel way by evaluating how the density of accesses compares against real-world effective memory bandwidths (measured by STREAM) and how it can be compared across widely varying architectures including GPUs and x86, ARM, and Power CPUs. We demonstrate how Spatter can be used to generate analysis plots comparing different architectures and show that current GPU systems achieve up to 65% of STREAM bandwidth for sparse accesses and are more energy efficient in doing so for several different sparsity patterns. Our future plans for the spatter benchmark are to use these results to predict the impact of new memory access primitives on various architectures, develop backends for novel hardware like FPGAs and the Emu Chick, and automate testing so that users can perform their own sparse access studies.
DOI：
--
发表时间：
2018
期刊：
IEEE International Parallel and Distributed Processing Symposium Workshops
影响因子：
--
作者：
Hein, Eric;Conte, Tom;Young, Jeffrey S.;Eswar, Srinivas;Li, Jiajia;Lavin, Patrick;Vuduc, Richard;Riedy, Jason
通讯作者：
Riedy, Jason