Efficient handling of data structures in definitional languages

Szymański, Bolesław K.; Prywes, Noah S.

doi:10.1016/0167-6423(88)90050-0

Cited by 8 publications

(5 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fortunately, analysis of array assignments and references can often show that each iteration step through a loop uses only a fixed number of adjacent elements of a data structure. This allows the object code to allocate a "window" large enough for only this fixed number of elements, not the entire dimension of the structure [13]. As with the original FORTRAN, the elements in the window are overwritten in each iteration.…”

Section: Figure 1 Overview Of Epl Systemmentioning

confidence: 99%

Integrating Data and Task Parallelism in Scientific Programs

Deelman

Kaplow

Szymański

et al. 1996

Languages, Compilers and Run-Time Systems for Scalable Computers

Self Cite

View full text Add to dashboard Cite

Functional languages attract the attention of developers of parallelizing compilers because of the implicit parallelism of functional programs and the simplified data dependence analysis of functional statements. A major drawback of functional languages is that naive translation of functional programs results in code that requires excessive memory. In this paper we explore the connection between the memory optimization and communication optimization of parallel codes generated from functional languages. We also show how a functional language can be used as an intermediate form in the translation from FORTRAN to customized, architecture-specific parallel code.

show abstract

Section: Figure 1 Overview Of Epl Systemmentioning

confidence: 99%

Integrating Data and Task Parallelism in Scientific Programs

Deelman

Kaplow

Szymański

et al. 1996

Languages, Compilers and Run-Time Systems for Scalable Computers

Self Cite

View full text Add to dashboard Cite

show abstract

“…scheduling and code generation for indi-vidual processes, memory optimization is performed to replace entire dimensions by windows of a few elements for multidimensional variables [7]. This step restores the memory efficiency of the original program.…”

Section: Additional Dimension Elimination: Duringmentioning

confidence: 99%

“…In the case of functional programs with single assignment and recurrent relations, the compile-time data-driven scheduling is decidable [26]. Such a scheduler has been implemented in the compiler for EPL language [7] and is not discussed here. Wavefront determination is presented below.…”

Section: In]mentioning

confidence: 99%

“…This extra temporal dimension allows the program to be specified without any reassignments but, unless optimized, may require an exorbitant amount of memory. The EPL compiler can often reduce the memory requirement of a program by replacing the entire dimension of an array by a few elements [7]. However, Sinharoy and Szymanski [27] have proven that the problem of finding the optimum replacement is equivalent to the well-known NP-hard problem of determining the maximum weight clique problem.…”

Section: In]mentioning

confidence: 99%

“…This article is intended as an overview of the research done towards implementing software tools as envisioned in Figure 1. More technical discussion can be found elsewhere [3][4][5][6][7][8][9][10].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Compiler Technology for Parallel Scientific Computation

Özturan

Sinharoy

Szymański

1994

Scientific Programming

Self Cite

View full text Add to dashboard Cite

There is a need for compiler technology that, given the source program, will generate efficient parallel codes for different architectures with minimal user involvement. Parallel computation is becoming indispensable in solving large-scale problems in science and engineering. Yet, the use of parallel computation is limited by the high costs of developing the needed software. To overcome this difficulty we advocate a comprehensive approach to the development of scalable architecture-independent software for scientific computation based on our experience with equational programming language (EPL). Our approach is based on a program decomposition, parallel code synthesis, and run-time support for parallel scientific computation. The program decomposition is guided by the source program annotations provided by the user. The synthesis of parallel code is based on configurations that describe the overall computation as a set of interacting components. Run-time support is provided by the compiler-generated code that redistributes computation and data during object program execution. The generated parallel code is optimized using techniques of data alignment, operator placement, wavefront determination, and memory optimization. In this article we discuss annotations, configurations, parallel code generation, and run-time support suitable for parallel programs written in the functional parallel programming language EPL and in Fortran.

show abstract

Scalable Software Tools for Parallel Computations

Szymański

1993

Software for Parallel Computation

View full text Add to dashboard Cite

Efficient handling of data structures in definitional languages

Cited by 8 publications

References 25 publications

Integrating Data and Task Parallelism in Scientific Programs

Integrating Data and Task Parallelism in Scientific Programs

Compiler Technology for Parallel Scientific Computation

Scalable Software Tools for Parallel Computations

Contact Info

Product

Resources

About