Models of parallel applications with large computation and I/O requirements

IEEE Transactions on Software Engineering

Volumn 28, Issue 3, 2002, Pages 286-307

  Rosti, Emilia ^a,b   Serazzi, Giuseppe ^a,c   Smirni, Evgenia ^a,d   Squillante, Mark S ^a,e  

^a IEEE

^b UNIVERSITY OF MILAN   (Italy)

^c POLITECNICO DI MILANO   (Italy)

^d The College of William and Mary   (United States)

^e IBM T J WATSON RESEARCH CENTER   (United States)

Author keywords

Parallel programs; Performance modeling and analysis; Performance prediction; Workload characterization

Indexed keywords

PARALLEL PROGRAMS; WORKLOAD CHARACTERIZATION;

ALGORITHMS; COMPUTATIONAL COMPLEXITY; INPUT OUTPUT PROGRAMS; OPTIMIZATION;

PARALLEL PROCESSING SYSTEMS;

EID: 0036504721     PISSN: 00985589     EISSN: None     Source Type: Journal    
DOI: 10.1109/32.991321     Document Type: Article

References (53)

1. Tuning the performance of I/O intensive parallel applications
2. Validity of the single processor approach to achieving large scale computing capabilities
3. Phase-type distributions and related point processes: Fitting and recent advances
4. Parallel I/O workload characteristics using vesta
5. The distributional little's law and its applications
6. A generalization of L = λW to moments of queue length and waiting times
7. Parallel computation of electron-molecule collisions
8. Input/output characterization of scalable parallel applications
9. On modeling partitioned multiprocessor systems
10. Speedup versus efficiency in parallel systems
11. Fitting mixtures of exponentials to long-tail distributions to analyze network performance models
12. Finite birth-and-death models in randomly changing environments
13. PPFS: A high performance portable parallel file system
14. Measuring parallel processor performance
15. Dynamic file-access characteristics of a production parallel scientific workload
16. Parameter approximation for phase-type distributions
17. Intelligent, adaptive file system policy selection
18. Input/output behavior of supercomputer applications
19. On some associative, parallel and analog computations
20. A performance evaluation of a general parallel processing model
21. File-eccess characteristics of parallel scientific workloads
22. A static analysis of I/O characteristics of scientific applications in a production workload
23. Dynamic I/O characterization of I/O intensive scientific applications
24. Characterizing parallel file-access patterns on a large-scale multiprocessor
25. Scalable performance analysis: The pablo performance analysis environment
26. The next frontier: Interactive and closed loop performance steering
27. The impact of I/O on program behavior and parallel scheduling
28. Application scheduling and processor allocation in multiprogrammed parallel processing systems
29. A comparison of logical and physical parallel I/O patterns
30. The EASY-loadLeveler API project
31. I/O requirements of scientific applications: An evolutionary view
32. Algorithmic influences on I/O access patterns and parallel file system performance
33. Lesson from characterizing the input/output behavior of parallel scientific applications
34. Analysis techniques for queueing network models of multi-computer systems with shared resources
35. The design and implementation of SOLAR, a portable library for scalable out-of-core linear algebra computations
36. A review of L = λW and extensions
37. The HP AutoRAID hierarchical storage system
38. Quantum chemical reaction dynamics on a highly parallel supercomputer