spECK: accelerating GPU sparse matrix-matrix multiplication through lightweight analysis
spECK: accelerating GPU sparse matrix-matrix multiplication through lightweight analysis
复制标题
spECK：通过轻量级分析加速 GPU 稀疏矩阵-矩阵乘法
DOI：
10.1145/3332466.3374521
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
M. Steinberger
中科院分区：
文献类型：
--
作者：
Mathias Parger;Martin Winter;Daniel Mlakar;M. Steinberger
由于稀疏矩阵的稀疏模式不同，因此GPU上的稀疏一般矩阵-矩阵乘法具有挑战性。现有的解决方案对于某些类型的矩阵实现了良好的性能，但是不能以相同的方式加速所有类型的矩阵。我们的方法结合了多种策略和动态参数选择，为矩阵的每一行动态选择和调整最佳拟合算法。这种选择得到了轻量级、多级矩阵分析的支持，它仔细平衡了分析成本和预期的性能增益。我们对数千种具有各种特性的矩阵进行了评估，结果表明，在所有超过15 k个产品的矩阵中，我们的性能优于79%的所有当前可用的解决方案，并且我们在15%的情况下实现了第二好的性能。对于这些矩阵，我们的解决方案比第二好的方法平均快83%，比其他最先进的GPU实现快25倍。使用我们的方法，应用程序可以预期很大的性能独立的矩阵，他们的工作。
Sparse general matrix-matrix multiplication on GPUs is challenging due to the varying sparsity patterns of sparse matrices. Existing solutions achieve good performance for certain types of matrices, but fail to accelerate all kinds of matrices in the same manner. Our approach combines multiple strategies with dynamic parameter selection to dynamically choose and tune the best fitting algorithm for each row of the matrix. This choice is supported by a lightweight, multi-level matrix analysis, which carefully balances analysis cost and expected performance gains. Our evaluation on thousands of matrices with various characteristics shows that we outperform all currently available solutions in 79% over all matrices with >15k products and that we achieve the second best performance in 15%. For these matrices, our solution is on average 83% faster than the second best approach and up to 25X faster than other state-of-the-art GPU implementations. Using our approach, applications can expect great performance independent of the matrices they work on.
DOI：
10.1109/hpec.2016.7761646
发表时间：
2016-06
期刊：
2016 IEEE High Performance Extreme Computing Conference (HPEC)
影响因子：
--
作者：
J. Kepner;Peter Aaltonen;David A. Bader;A. Buluç;F. Franchetti;J. Gilbert;D. Hutchison;Manoj Kumar;A. Lumsdaine;Henning Meyerhenke;Scott McMillan;Carl Yang;John Douglas Owens;Marcin Zalewski;T. Mattson;J. Moreira
通讯作者：
J. Kepner;Peter Aaltonen;David A. Bader;A. Buluç;F. Franchetti;J. Gilbert;D. Hutchison;Manoj Kumar;A. Lumsdaine;Henning Meyerhenke;Scott McMillan;Carl Yang;John Douglas Owens;Marcin Zalewski;T. Mattson;J. Moreira