Stochastic prediction of execution time for dynamic bulk synchronous computations

Journal of Supercomputing

Volumn 21, Issue 1, 2002, Pages 91-103

  Xu, Cheng Zhong ^a   Wang, Le Yi ^a   Fong, Ngo Tai ^a  

^a Wayne State University   (United States)

Author keywords

Bulk synchronous computation; Prediction of execution time; Stochastic analysis; Task scheduling

Indexed keywords

COMPUTATIONAL COMPLEXITY; COMPUTER SIMULATION; MATHEMATICAL MODELS; RANDOM PROCESSES; RESOURCE ALLOCATION; SCHEDULING;

BULK SYNCHRONOUS COMPUTATION; STOCHASTIC ANALYSIS; TASK SCHEDULING;

MULTIPROCESSING SYSTEMS;

EID: 0036173703     PISSN: 09208542     EISSN: None     Source Type: Journal    
DOI: 10.1023/A:1013539532035     Document Type: Article

References (19)

1. The relative performance of various mapping algorithms is independent of sizable variances in run-time predictions
2. A hierarchical approach for bounding the completion time distribution of stochastic task graphs
3. Optimal periodic remapping of bulk synchronous computations on multiprogrammed distributed systems
4. Run-time statistical estimation of task execution times for heterogeneous distributed computing
5. Statistical prediction of task execution times through analytic benchmarking for scheduling in a heterogeneous environment
6. Studies in the useful predictability of programs in a distributed and homogeneous environment
7. Stochastic bounds for parallel program execution times with processor constraints
8. Performance of synchronous parallel algorithms with regular structures
9. Optimal dynamic remapping of data parallel computation
10. Dynamic remapping of parallel computations with varying resource demands
11. Beyond execution time: Expanding the use of performance models
12. Static dependent costs for estimating execution time
13. Some analysis techniques for asynchronous multiprocessor algorithms
14. Estimation of execution times on heterogeneous supercomputer architecture