Hybrid static/dynamic scheduling for already optimized dense matrix factorization

被引:9
|
作者
Donfack, Simplice [1 ]
Grigori, Laura [1 ]
Gropp, William D. [2 ]
Kale, Vivek [2 ]
机构
[1] Univ Paris 11, INRIA Saclay Ile France, Bat 425, F-91405 Orsay, France
[2] Univ Illinois, Dept Comp Sci, Urbana, IL 61801 USA
关键词
dynamic scheduling; communication-avoiding; LU factorization; numerical linear algebra; LOCALITY;
D O I
10.1109/IPDPS.2012.53
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
We present the use of a hybrid static/dynamic scheduling strategy of the task dependency graph for direct methods used in dense numerical linear algebra. This strategy provides a balance of data locality, load balance, and low dequeue overhead. We show that the usage of this scheduling in communication avoiding dense factorization leads to significant performance gains. On a 48 core AMD Opteron NUMA machine, our experiments show that we can achieve up to 64% improvement over a version of CALU that uses fully dynamic scheduling, and up to 30% improvement over the version of CALU that uses fully static scheduling. On a 16-core Intel Xeon machine, our hybrid static/dynamic scheduling approach is up to 8% faster than the version of CALU that uses a fully static scheduling or fully dynamic scheduling. Our algorithm leads to speedups over the corresponding routines for computing LU factorization in well known libraries. On the 48 core AMD NUMA machine, our best implementation is up to 110% faster than MKL, while on the 16 core Intel Xeon machine, it is up to 82% faster than MKL. Our approach also shows significant speedups compared with PLASMA on both of these systems.
引用
收藏
页码:496 / 507
页数:12
相关论文
共 50 条
  • [21] MATRIX FACTORIZATION METHOD IN MIXED STATIC PROBLEMS OF AN ELASTIC WEDGE
    BERKOVICH, VN
    PMM JOURNAL OF APPLIED MATHEMATICS AND MECHANICS, 1976, 40 (04): : 623 - 631
  • [22] Hybrid evolutionary static scheduling for heterogeneous systems
    Boeres, C
    Rios, E
    Ochi, LS
    2005 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION, VOLS 1-3, PROCEEDINGS, 2005, : 1929 - 1936
  • [23] Load-balancing scheduling of simulation tasks based on a static-dynamic hybrid algorithm
    Wang, Xiashuang
    Li, Ni
    Gong, Guanghong
    Song, Xiao
    Guo, Yanqi
    JOURNAL OF SIMULATION, 2022, 16 (02) : 182 - 193
  • [24] A hybrid Static/Dynamic DVS scheduling for real-time systems with (m,k)-guarantee
    Niu, LW
    Quan, G
    RTSS 2005: 26TH IEEE INTERNATIONAL REAL-TIME SYSTEMS SYMPOSIUM, PROCEEDINGS, 2005, : 356 - 365
  • [25] HYBRID MATRIX FACTORIZATION FOR RECOMMENDER SYSTEMS IN SOCIAL NETWORKS
    Zhao, C.
    Sun, S.
    Han, L.
    Peng, Q.
    NEURAL NETWORK WORLD, 2016, 26 (06) : 559 - 569
  • [26] Hybrid Matrix Factorization for Multi-view Clustering
    Yu, Hongbin
    Shu, Xin
    INTELLIGENCE SCIENCE AND BIG DATA ENGINEERING: BIG DATA AND MACHINE LEARNING, PT II, 2019, 11936 : 302 - 311
  • [27] CROSS-MODAL RETRIEVAL BASED ON OPTIMIZED MATRIX FACTORIZATION HASHING
    Wu, Wei
    Jing, Xiaoyuan
    Du, Wencai
    Cao, Xinghui
    Zhou, Hui
    JOURNAL OF NONLINEAR AND CONVEX ANALYSIS, 2019, 20 (06) : 1173 - 1181
  • [28] Optimized Dense Matrix Multiplication on a Many-Core Architecture
    Garcia, Elkin
    Venetis, Ioannis E.
    Khan, Rishi
    Gao, Guang R.
    EURO-PAR 2010 - PARALLEL PROCESSING, PART II, 2010, 6272 : 316 - +
  • [29] Static and dynamic stability in dense medium separation processes
    Univ of Trieste, Trieste, Italy
    Mineral Processing and Extractive Metallurgy Review, 2000, 20 (01): : 197 - 214
  • [30] Bayesian Dynamic Mode Decomposition with Variational Matrix Factorization
    Kawashima, Takahiro
    Shouno, Hayaru
    Hino, Hideitsu
    THIRTY-FIFTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THIRTY-THIRD CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE AND THE ELEVENTH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2021, 35 : 8083 - 8091