Title :
A performance evaluation of list scheduling heuristics for task graphs without communication costs
Author :
Brest, Janez ; Zumer, Viljem
Author_Institution :
Fac. of Electr. Eng. & Comput. Sci., Maribor Univ., Slovenia
Abstract :
Proposes three new static scheduling algorithms for allocating task graphs without communication costs to fully connected multiprocessors. The proposed algorithms, called MCP/ABS, MCP/CLR and MCP/CLRR, are based on the well-known MCP (modified critical path) algorithm, and all of them have the same complexity of O(v2log v), where v is the number of nodes in the task graph. A global comparison of the proposed algorithms with three recently reported scheduling algorithms [ETF (earliest task first), HLFET (high level first with estimated time) and MCP] is carried out. The proposed algorithms generate similar or even better solutions than the previous algorithms in terms of the completion times of the resulting schedules using a prototype standard task graph set
Keywords :
computational complexity; directed graphs; list processing; parallel algorithms; processor scheduling; software performance evaluation; ETF algorithm; HLFET algorithm; MCP algorithm; MCP/ABS algorithm; MCP/CLR algorithm; MCP/CLRR algorithm; communication costs; computational complexity; fully connected multiprocessors; list scheduling heuristics; modified critical path algorithm; performance evaluation; prototype standard task graph set; schedule completion times; static scheduling algorithms; task graph allocation; task graph nodes; Clustering algorithms; Costs; Dynamic scheduling; Heuristic algorithms; Multiprocessing systems; NP-hard problem; Optimal scheduling; Processor scheduling; Prototypes; Scheduling algorithm;
Conference_Titel :
Parallel Processing, 2000. Proceedings. 2000 International Workshops on
Conference_Location :
Toronto, Ont.
Print_ISBN :
0-7695-0771-9
DOI :
10.1109/ICPPW.2000.869147