Scheduling tree-shaped task graphs to minimize memory and makespan

被引:5
|
作者
Marchal, Loris [1 ,2 ]
Sinnen, Oliver [3 ]
Vivien, Frederic [2 ,4 ]
机构
[1] CNRS, Lyon, France
[2] Univ Lyon, Lyon, France
[3] Univ Auckland, Auckland, New Zealand
[4] INRIA, Lyon, France
来源
IEEE 27TH INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM (IPDPS 2013) | 2013年
关键词
scheduling; makespan-memory tradeoff; tree-shaped task-graphs;
D O I
10.1109/IPDPS.2013.55
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
This paper investigates the execution of tree-shaped task graphs using multiple processors. Each edge of such a tree represents a large IO file. A task can only be executed if all input and output files fit into memory, and a file can only be removed from memory after it has been consumed. Such trees arise, for instance, in the multifrontal method of sparse matrix factorization. The maximum amount of memory needed depends on the execution order of the tasks. With one processor the objective of the tree traversal is to minimize the required memory. This problem was well studied and optimal polynomial algorithms were proposed. Here, we extend the problem by considering multiple processors, which is of obvious interest in the application area of matrix factorization. With the multiple processors comes the additional objective to minimize the time needed to traverse the tree, i.e., to minimize the makespan. Not surprisingly, this problem proves to be much harder than the sequential one. We study the computational complexity of this problem and provide an inapproximability result even for unit weight trees. Several heuristics are proposed, each with a different optimization focus, and they are analyzed in an extensive experimental evaluation using realistic trees.
引用
收藏
页码:839 / 850
页数:12
相关论文
共 50 条
  • [1] PRACTICAL STEADY-STATE SCHEDULING FOR TREE-SHAPED TASK GRAPHS
    Diakite, Sekou
    Marchal, Loris
    Nicod, Jean-Marc
    Philippe, Laurent
    PARALLEL PROCESSING LETTERS, 2011, 21 (04) : 397 - 412
  • [2] Partitioning Tree-Shaped Task Graphs for Distributed Platforms With Limited Memory
    Gou, Changjiang
    Benoit, Anne
    Marchal, Loris
    IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2020, 31 (07) : 1533 - 1544
  • [3] Scheduling series-parallel task graphs to minimize peak memory
    Kayaaslan, Enver
    Lambert, Thomas
    Marchal, Loris
    Ucar, Bora
    THEORETICAL COMPUTER SCIENCE, 2018, 707 : 1 - 23
  • [4] RRTS: A Task Scheduling Algorithm to Minimize Makespan in Grid Environment
    Panda, Sanjaya Kumar
    Bhoi, Sourav Kumar
    Khilar, Pabitra Mohan
    PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INTERNET COMPUTING AND INFORMATION COMMUNICATIONS (ICICIC GLOBAL 2012), 2014, 216 : 279 - 292
  • [5] Context-aware scheduling analysis of distributed systems with tree-shaped task-dependencies
    Henia, R
    Ernst, R
    DESIGN, AUTOMATION AND TEST IN EUROPE CONFERENCE AND EXHIBITION, VOLS 1 AND 2, PROCEEDINGS, 2005, : 480 - 485
  • [6] A genetic algorithm for scheduling open shops with conflict graphs to minimize the makespan
    Tellache, Nour ElHouda
    Kerbache, Laoucine
    COMPUTERS & OPERATIONS RESEARCH, 2023, 156
  • [7] Scheduling with Rejection to Minimize the Makespan
    Zhang, Yuzhong
    Ren, Jianfeng
    Wang, Chengfei
    COMBINATORIAL OPTIMIZATION AND APPLICATIONS, PROCEEDINGS, 2009, 5573 : 411 - 420
  • [8] Algorithms for tree-shaped task partition and allocation on heterogeneous multiprocessors
    Suna He
    Jigang Wu
    Bing Wei
    Jiaxin Wu
    The Journal of Supercomputing, 2023, 79 : 13210 - 13240
  • [9] Algorithms for tree-shaped task partition and allocation on heterogeneous multiprocessors
    He, Suna
    Wu, Jigang
    Wei, Bing
    Wu, Jiaxin
    JOURNAL OF SUPERCOMPUTING, 2023, 79 (12): : 13210 - 13240
  • [10] Performance evaluation of task scheduling algorithms in virtual cloud environment to minimize makespan
    Kaur R.
    Laxmi V.
    Balkrishan
    International Journal of Information Technology, 2022, 14 (1) : 79 - 93