We have studied the effectiveness of parallelizing compilers and the underlying transformation techniques. This paper reports the speedups of the Perfect Benchmarks TM codes that result from automatic parallelization. We have further measured the performance gains caused by individual restructuring techniques. Specific reasons for the successes and failures of the transformations are discussed, and potential improvements that result in measurably better program performance are analyzed. Our most important findings are that available restructurers often cause insignificant performance gains in real programs and that only few restructuring techniques contribute to this gain. However, we can also show that there is potential for advancing compiler technology so that many of the most important loops in these programs can be parallelized.
|
676
|
A data locality optimizing algorithm
– Wolf, Lam
- 1991
|
|
296
|
Advanced compiler optimizations for supercomputers
– Padua, Wolfe
- 1986
|
|
213
|
The Perfect Club Benchmarks: Effective Performance Evaluation of Supercomputers
– Berry, Chen, et al.
- 1989
|
|
204
|
Supernode partitioning
– Irigoin, Triolet
- 1988
|
|
173
|
More iteration space tiling
– Wolfe
- 1989
|
|
169
|
Scanning polyhedra with DO loops
– Ancourt, Irigoin
- 1991
|
|
137
|
Practical dependence testing
– Goff, Kennedy, et al.
- 1991
|
|
110
|
Efficient and exact data dependence analysis
– Maydan, Hennessy, et al.
|
|
105
|
Optimal loop parallelization
– Aiken, Nicolau
- 1988
|
|
102
|
Experience in the automatic parallelization of four perfect benchmark programs
– Hoeflinger, Li, et al.
- 1992
|
|
80
|
Compiler algorithms for synchronization
– Midkiff, Padua
- 1987
|
|
59
|
Runtime compilation methods for multicomputers
– Wu, Saltz, et al.
- 1991
|
|
26
|
An Effectiveness Study of Parallelizing Compiler Techniques
– Eigenmann, Blume
- 1991
|
|
24
|
The PERFECT club benchmarks: E ective performance evaluation of supercomputers
– Sameh, Clementi, et al.
- 1989
|
|
23
|
Vectorizing Compilers: A Test Suite and Results
– Callahan, Dongarra, et al.
- 1988
|
|
18
|
On reducing data synchronization in multiprocessed loops
– Li, Abu-Sufah
- 1987
|
|
16
|
Machine-Independent Evaluation of Parallelizing Compilers
– Petersen, Padua
- 1992
|
|
15
|
An Evaluation of Automatic and Interactive Parallel Programming Tools
– Cheng, Pase
- 1991
|
|
14
|
and Pen-Chung Yew. A Scheme to Enforce Data Dependence on Large Multiprocessor Systems
– Zhu
- 1987
|
|
12
|
Automatic Recognition of Induction & Recurrence Relations by Abstract Interpretation
– Ammarguellat, Harrison
- 1990
|
|
11
|
Removal of redundant dependences in DOACROSS loops with constant dependences
– Krothapalli, Sadayappan
- 1991
|
|
8
|
Performance Evaluation of three Automatic Vectorizer Packages
– Arnold
- 1982
|
|
7
|
The Effect of Restructuring Compilers on Program Performance for High-Speed Computers
– Cytron, Kuck, et al.
- 1985
|
|
6
|
Cedar Fortran and its Restructuring Compiler
– Eigenmann, Hoeflinger, et al.
- 1990
|
|
6
|
A Comparison Study of Automatically Vectorizing Fortran Compilers
– Nobayashi, Eoyang
- 1989
|
|
5
|
Optimization of data/control conditions in task graphs
– Girkar, Polychronopoulos
- 1992
|
|
4
|
An Evaluation of Vector FORTRAN 200 Generated by CYBER 205 and ETA-10 Pre-Compilation Tools
– Braswell, Keech
- 1988
|
|
3
|
Programmiertechniken fur die Vektorisierung
– Detert
- 1987
|
|
3
|
Homogeneous Boolean Algebras may have non-simple automorphism groups
– Kasahara, Honda, et al.
- 1990
|
|
3
|
A comparative study of KAP and VAST: two automatic preprocessors with Fortran 8x Output. Supercomputer 28
– Luecke, Coyle, et al.
- 1988
|
|
2
|
Restructuring Fortran Programs for Cedar. to appear in Concurrency: Practice and Experience
– Eigenmann, Hoeflinger, et al.
- 1992
|
|
2
|
QCD Optimization Report
– Hoeflinger
- 1991
|
|
1
|
GTS: Parrallelization and vectorization of tight recurrences
– Ayguad'e, Labarta, et al.
|
|
1
|
QCD Optimization Report
– inger
- 1991
|
|
1
|
A compilation scheme for macro-data ow computation on hierarchical multiprocessor systems
– Kasahara, Honda, et al.
- 1990
|
|
