An efficient mixed-mode representation of sparse tensors
An efficient mixed-mode representation of sparse tensors
复制标题
DOI：
10.1145/3295500.3356216
复制
发表时间：
2019-11
期刊：
影响因子：
--
通讯作者：
Israt Nisa;Jiajia Li;Aravind Sukumaran-Rajam;Prasant Singh;Sri-ram Krishnamoorthy;P. Sadayappan;Singh Rawat;Sri-ram Krishnamoorthy;An Efficient Mixed-Mode
中科院分区：
文献类型：
--
作者：
Israt Nisa;Jiajia Li;Aravind Sukumaran-Rajam;Prasant Singh;Sri-ram Krishnamoorthy;P. Sadayappan;Singh Rawat;Sri-ram Krishnamoorthy;An Efficient Mixed-Mode
稀疏张量的压缩稀疏纤维（CSF）表示是稀疏矩阵的压缩稀疏行（CSR）格式的推广。对于具有d个模式的张量，典型的张量方法（例如CANDECOMP/PARAFAC分解（CPD））需要一系列d个张量计算，其中每个张量计算都需要相对于不同模式的有效存储器访问。简单的解决方案是使用d个不同的张量表示，每个表示对于d个计算中的一个是有效的。然而，d倍空间开销在实践中通常是不可接受的，特别是对于内存受限的GPU。在本文中，我们提出了一种混合模式张量表示，张量的非零元素分割成不相交的部分，其中每一个被压缩，以创建纤维沿着不同的模式。实验结果表明，可以实现更好的性能，同时只利用一小部分的空间，需要保持d不同的CSF表示。
The Compressed Sparse Fiber (CSF) representation for sparse tensors is a generalization of the Compressed Sparse Row (CSR) format for sparse matrices. For a tensor with d modes, typical tensor methods such as CANDECOMP/PARAFAC decomposition (CPD) require a sequence of d tensor computations, where efficient memory access with respect to different modes is required for each of them. The straightforward solution is to use d distinct representations of the tensor, with each one being efficient for one of the d computations. However, a d-fold space overhead is often unacceptable in practice, especially with memory-constrained GPUs. In this paper, we present a mixed-mode tensor representation that partitions the tensor's nonzero elements into disjoint sections, each of which is compressed to create fibers along a different mode. Experimental results demonstrate that better performance can be achieved while utilizing only a small fraction of the space required to keep d distinct CSF representations.