Scheduling OR-parallelism in YapOr and ThOr on Multi-Core Machines

被引:0
|
作者
Dutra, Ines [1 ,2 ]
Rocha, Ricardo [1 ,2 ]
Costa, Vitor Santos [1 ,2 ]
Silva, Fernando [1 ,2 ]
Santos, Joao [1 ,2 ]
机构
[1] Univ Porto, CRACS, Rua Campo Alegre 1021-1055, P-4169007 Oporto, Portugal
[2] Univ Porto, INESC Porto TEC, Fac Sci, Dept Comp Sci, P-4169007 Oporto, Portugal
关键词
Scheduling strategies; Parallelism; Prolog implementation; PROLOG; AURORA; MUSE; EXECUTION; PROGRAMS;
D O I
10.1109/IPDPSW.2012.199
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
In this work we perform a detailed study of different or-scheduling strategies varying several parameters in two or-parallel systems, YapOr and ThOr, running on multi-core machines. Our results show that some kinds of applications are sensitive to the choice of scheduling strategy adopted. In particular, the choice of scheduling parameters mostly affect applications that have short execution times, which, despite having speedups, have their performance significantly affected. Our results also show that topmost dispatching can be more advantageous than bottommost dispatching, a finding that contradicts previous works in this area. One last finding is that YapOr and ThOr are affected differently by changes in scheduling with ThOr performing significantly better than YapOr in several applications.
引用
收藏
页码:1581 / 1590
页数:10
相关论文
共 50 条
  • [1] Improved parallelism and scheduling in multi-core software routers
    Egi, Norbert
    Iannaccone, Gianluca
    Manesh, Maziar
    Mathy, Laurent
    Ratnasamy, Sylvia
    JOURNAL OF SUPERCOMPUTING, 2013, 63 (01): : 294 - 322
  • [2] Improved parallelism and scheduling in multi-core software routers
    Norbert Egi
    Gianluca Iannaccone
    Maziar Manesh
    Laurent Mathy
    Sylvia Ratnasamy
    The Journal of Supercomputing, 2013, 63 : 294 - 322
  • [3] Efficiently Scheduling Multi-core Guest Virtual Machines on Multi-core Hosts in Network Simulation
    Yoginath, Srikanth B.
    Perumalla, Kalyan S.
    2011 IEEE WORKSHOP ON PRINCIPLES OF ADVANCED AND DISTRIBUTED SIMULATION (PADS), 2011,
  • [4] FULL PROLOG AND SCHEDULING OR-PARALLELISM IN MUSE
    ALI, KAM
    KARLSSON, R
    INTERNATIONAL JOURNAL OF PARALLEL PROGRAMMING, 1990, 19 (06) : 445 - 475
  • [5] DAG Scheduling and Analysis on Multi-Core Systems by Modelling Parallelism and Dependency
    Zhao, Shuai
    Dai, Xiaotian
    Bate, Iain
    IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2022, 33 (12) : 4019 - 4038
  • [6] Scheduling tasks with precedence constraints on hybrid multi-core machines
    Kedad-Sidhoum, Safia
    Monna, Florence
    Trystram, Denis
    2015 IEEE 29TH INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM WORKSHOPS, 2015, : 27 - 33
  • [7] Generic Algorithms for Scheduling Applications on Hybrid Multi-core Machines
    Amaris, Marcos
    Lucarelli, Giorgio
    Mommessin, Clement
    Trystram, Denis
    EURO-PAR 2017: PARALLEL PROCESSING, 2017, 10417 : 220 - 231
  • [8] Runtime scheduling of dynamic parallelism on accelerator-based multi-core systems
    Blagojevic, Filip
    Nikolopoulos, Dimitrios S.
    Stamatakis, Alexandros
    Antonopoulos, Christos D.
    Curtis-Maury, Matthew
    PARALLEL COMPUTING, 2007, 33 (10-11) : 700 - 719
  • [9] Combining SIMD and Many/Multi-core Parallelism for Finite State Machines with Enumerative Speculation
    Jiang, Peng
    Agrawal, Gagan
    ACM SIGPLAN NOTICES, 2017, 52 (08) : 179 - 191
  • [10] FLEXIBLE SCHEDULING OF OR-PARALLELISM IN AURORA - THE BRISTOL SCHEDULER
    BEAUMONT, A
    RAMAN, SM
    SZEREDI, P
    WARREN, DHD
    LECTURE NOTES IN COMPUTER SCIENCE, 1991, 506 : 403 - 420