Efficient parallel execution of irregular recursive programs

IEEE Transactions on Parallel and Distributed Systems

Volumn 13, Issue 2, 2002, Pages 167-178

  Prechelt, Lutz ^a,b   Hänßgen, Stefan U ^c  

^a IEEE   (Germany)

^b abaXX Technology   (Germany)

^c GINIT ^*

Author keywords

Benchmarks; Granularity control; Instrumentation; Irregular problems; Profiling; Recursion; SMP

Indexed keywords

GRANULARITY CONTROL APPROACH; RECURSIONS AUTOMATICALLY PARALLELIZED; SOFTWARE PACKAGE SOLARIS; SYMMETRIC MULTIPROCESSOR MACHINES;

C (PROGRAMMING LANGUAGE); COMPUTER OPERATING SYSTEMS; COMPUTER SOFTWARE SELECTION AND EVALUATION; MACHINE ORIENTED LANGUAGES; MULTIPROCESSING PROGRAMS; PROGRAM COMPILERS; RESPONSE TIME (COMPUTER SYSTEMS);

PARALLEL PROCESSING SYSTEMS;

EID: 0036472469     PISSN: 10459219     EISSN: None     Source Type: Journal    
DOI: 10.1109/71.983944     Document Type: Article

References (24)

1. A hierarchical O(N log N) force calculation algorithm
2. Bitonic sorting
3. Programming parallel algorithms
4. Concurrency and distribution in object-oriented programming
5. Retire fortran? A debate rekindled
6. Olden: Parallelizing programs with dynamic data structures on distributed-memory machines
7. Randomized load balancing for tree-structured computation
8. Parallel programming in split-C
9. Using the P3T to guide the parallelization and optimization effort under the vienna fortran compilation system
10. High performance fortran forum
11. Effiziente parallel ausführung irregulärer rekursiver programme
12. REAPAR User manual and reference: Automatic parallelization of irregular recursive programs
13. Compiling fortran D for MIMD distributed memory machines
14. Optimizing dynamically-dispatched calls with run-time type feedback
15. Programming frames for the efficient use of parallel systems
16. The cilk system for parallel multithreaded computing
17. Automatic alignment of array data and processes to reduce communication time on DMPPs
18. Imperative concurrent object-oriented languages
19. Object-oriented divide-and-conquer for parallel processing