Regular distributions for storing dense matrices on parallel systems are not always used in practice. In many scientific applicati RUMMA) [1] to handle irregularly distributed matrices. Our approach relies on a distribution independent algorithm that provides dynamic load balancing by exploiting data locality and achieves performance as good as the traditional approach which relies on temporary arrays with regular distribution, data redistribution, and matrix multiplication for regular matrices to handle the irregular case. The proposed algorithm is memory-efficient because temporary matrices are not needed. This feature is critical for systems like the IBM Blue Gene/L that offer very limited amount of memory per node. The experimental results demonstrate very good performance across the range of matrix distributions and problem sizes motivated by real applications.
展开▼
机译：在实践中并不总是使用用于在并行系统上存储密集矩阵的规则分布。在许多科学应用中，RUMMA）[1]处理不规则分布的矩阵。我们的方法依赖于一种独立于分布的算法，该算法通过利用数据局部性来提供动态负载平衡，并获得了与传统方法相同的性能，传统方法依赖于具有规则分布的临时数组，数据重新分布以及规则矩阵的矩阵乘法来处理不规则情况。由于不需要临时矩阵，因此所提出的算法具有存储效率。对于像IBM Blue Gene / L这样的每个节点提供非常有限的内存量的系统而言，此功能至关重要。实验结果表明，在由实际应用引起的矩阵分布和问题大小范围内，它们都具有非常好的性能。
展开▼
