An overview of the PTRAN analysis system for multiprocessing

Allen, Frances E.; Burke, Michael G.; Charles, Philippe; Cytron, Ron K.; Ferrante, Jeanne

doi:10.1007/3-540-18991-2_12

Cited by 35 publications

(37 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This algorithm embodies two techniques for dealing with data dependences: The first (sequencing) satisfies such dependences implicitly by reducing parallelism, and the second (privatization) eliminates some dependences by introducing private variables. This algorithm has been implemented in the PTRAN system [Allen, E, et al 1988;Hsieh 1988]. …”

Section: Introductionmentioning

confidence: 99%

“…Such a representation has important advantages in the context of automatic parallelization [Allen, E, et al 1988;Baxter and Bauer 1989;Cytron et al 1989;Ferrante et al 1987]. Moreover, this representation is language independent: Analysis over such a representation can be applied to any imperative sequential programming language.…”

Section: Introductionmentioning

confidence: 99%

“…For input, we assume that a sequential procedure has been analyzed and its associated control and data dependence graphs have been constructed [Allen, E et al 1988]. Such analysis may incorporate the effects of other procedures or facts provided by an informed user.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automatic generation of nested, fork-join parallelism

et al. 1989

Self Cite

View full text Add to dashboard Cite

This paper presents an efficient algorithm that automatically generates a parallel program from a dependence-based representation of a sequential program. The resulti~ parallel program consists of nested fork-join constructs, composed from the loops and statements of the sequential program. Data dependences are handled b,, two techniques. One technique implicitly satisfies them b? sequencing, thereby reducing parallelism. Where increased parallelism results, the other technique eliminates them by privatization: the introduction of processspecific private instances of variables. Additionally, the algorithm determines when cowing values of such instances in and out of nested parallel constructs results in greater parallelism. This is the first algorithm for automatically generating parallelism for such a general model. The algorithm generates as much parallelism as is possible in our model while minimizing privatization.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automatic generation of nested, fork-join parallelism

et al. 1989

Self Cite

View full text Add to dashboard Cite

show abstract

“…Parallelizing compilers [Allen et al 1988;Polychronopoulos 1988;Polychronopoulos et al 1989;Wolfe 1989], like conventional optimizing compilers, collect data flow information for a source program and use this information to detect potential parallelism, determine an appropriate grain size, and then transform the program into a functionally equivalent program with a higher degree of parallelism. They can take hours to compile programs of reasonable size [Gross et al 1989].…”

Section: Applicationsmentioning

confidence: 99%

Experiences with a parallel algorithm for data flow analysis

1991

View full text Add to dashboard Cite

We have designed a family of parallel data flow analysis algorithms for execution on distributed-memory MIMD machines, based on general-purpose, hybrid algorithms for data flow analysis [Marlowe and Ryder 1990]. We exploit a natural partitioning of the hybrid algorithms and explore a static mapping, dynamic scheduling strategy. Alternative mapping-scheduling choices and refinements of the flow graph condensation used are discussed. Our parallel hybrid algorithm family is illustrated on Reaching Definitions, although parallel algorithms also exist for many interprocedural (e.g., Aliasing) and intraprocedural (e.g., Available Expressions) problems [Marlowe 1989]. We have implemented the parallel hybrid algorithm for Reaching Definitions on an Intel iPSC/2. Our empirical results suggest the practicality of parallel hybrid algorithms.

show abstract

“…Parallelizing compilers [1,8,12,19] rely upon subscript analysis [2,18] to detect data dependences between pairs of array references inside loop nests. The array data dependence problem is equivalent to integer programming, a well known NP-hard problem, and thus it can not be solved efficiently in general.…”

Section: Introductionmentioning

confidence: 99%

An Empirical Study of the I Test for Exact Data Dependence

Psarris

Pande

1994

1994 International Conference on Parallel Processing Vol. 3

View full text Add to dashboard Cite

Parallelizing Compilers rely upon subscript analysis to detect data dependences between pairs of array references inside loop nests. The most widely used approximate subscript analysis tests are the GCD test and the Banerjee test. In an earlier work we proposed the I test, an improved subscript analysis test. The I test extends the accuracy of a combination of the GCD test and the Banerjee test. It is also able to provide exact data dependence information at no additional computation cost. In the present work we perform an empirical study on the Perfect Club benchmarks to demonstrate the effectiveness and practical importance of the I Test. We compare its performance with that of the GCD test and the Banerjee test. We show that the I test is always an exact test in practice.

show abstract

An overview of the PTRAN analysis system for multiprocessing

Cited by 35 publications

References 21 publications

Automatic generation of nested, fork-join parallelism

Automatic generation of nested, fork-join parallelism

Experiences with a parallel algorithm for data flow analysis

An Empirical Study of the I Test for Exact Data Dependence

Contact Info

Product

Resources

About