Performance Evaluation of Sparse Matrix Multiplication Kernels on Intel Xeon Phi

Intel Xeon Phi is a recently released high-performance coprocessor which features 61 cores each supporting 4 hardware threads with 512-bit wide SIMD registers achieving a peak theoretical performance of 1Tflop/s in double precision. Its design differs from classical modern processors; it comes with a large number of cores, the 4-way hyperthreading capability allows many applications to saturate the massive memory bandwidth, and its large SIMD capabilities allow to reach high computation throughput. The core of many scientific applications involves the multiplication of a large, sparse matrix with a single or multiple dense vectors which are not compute-bound but memory-bound. In this paper, we investigate the performance of the Xeon Phi coprocessor for these sparse linear algebra kernels. We highlight the important hardware details and show that Xeon Phi’s sparse kernel performance is very promising and even better than that of cutting-edge CPUs and GPUs.

[1]  James Demmel,et al.  When cache blocking of sparse matrix vector multiply works and why , 2007, Applicable Algebra in Engineering, Communication and Computing.

[2]  Ankit Jain.  pOSKI : An Extensible Autotuning Framework to Perform Optimized SpMVs on Multicore Architectures , 2008 .

[3]  Zheng Zhou,et al.  An Out-of-Core Eigensolver on SSD-equipped Clusters , 2012, 2012 IEEE International Conference on Cluster Computing.

[4]  John R. Gilbert,et al.  Parallel sparse matrix-vector and matrix-transpose-vector multiplication using compressed sparse blocks , 2009, SPAA '09.

[5]  Louis-Noël Pouchet,et al.  Automatic Transformations for Effective Parallel Execution on Intel Many Integrated Core , 2012 .

[6]  Katherine A. Yelick,et al.  Optimizing Sparse Matrix Computations for Register Reuse in SPARSITY , 2001, International Conference on Computational Science.

[7]  D. Panda,et al.  Intra-MIC MPI Communication using MVAPICH 2 : Early Experience , 2012 .

[8]  Youcef Saad,et al.  A Basic Tool Kit for Sparse Matrix Computations , 1990 .

[9]  Samuel Williams,et al.  Reduced-Bandwidth Multithreaded Algorithms for Sparse Matrix-Vector Multiplication , 2011, 2011 IEEE International Parallel & Distributed Processing Symposium.

[10]  Ümit V. Çatalyürek,et al.  Fast Recommendation on Bibliographic Networks , 2012, 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining.

[11]  Ümit V. Çatalyürek,et al.  An Early Evaluation of the Scalability of Graph Algorithms on the Intel MIC Architecture , 2012, 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops & PhD Forum.

[12]  Michael Klemm,et al.  OpenMP Programming on Intel Xeon Phi Coprocessors: An Early Performance Comparison , 2012, MARC@RWTH.

[13]  Samuel Williams,et al.  Optimization of sparse matrix-vector multiplication on emerging multicore platforms , 2007, Proceedings of the 2007 ACM/IEEE Conference on Supercomputing (SC '07).

[14]  Marcin Dabrowski,et al.  Parallel symmetric sparse matrix-vector product on scalar multi-core CPUs , 2010, Parallel Comput..

[15]  E. Cuthill,et al.  Reducing the bandwidth of sparse symmetric matrices , 1969, ACM '69.

[16]  Michael Garland,et al.  Implementing sparse matrix-vector multiplication on throughput-oriented processors , 2009, Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis.

[17]  Katherine Yelick,et al.  OSKI: A library of automatically tuned sparse matrix kernels , 2005 .

[18]  John M. Mellor-Crummey,et al.  Optimizing Sparse Matrix–Vector Product Computations Using Unroll and Jam , 2004, Int. J. High Perform. Comput. Appl..