Compiling scientific code using partial evaluation

Berlin, Andrew A.; Weise, Daniel

doi:10.1109/2.62091

Cited by 61 publications

(29 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This strategy has been widely researched in the literature on interprocedural analysis [11,17] and in work on partial evaluation [3,25]. Our strategy differs from most of the previous work in that the specializations in a telescoping language must be generated before the calling program is revealed.…”

Section: Low-level Code Expansionmentioning

confidence: 77%

Telescoping Languages: A Strategy for Automatic Generation of Scientific Problem-Solving Systems from Annotated Libraries

Kennedy

Broom

Cooper

et al. 2001

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

As machines and programs have become more complex, the process of programming applications that can exploit the power of high-performance systems has become more difficult and correspondingly more labor-intensive. This has substantially widened the software gap the discrepancy between the need for new software and the aggregate capacity of the workforce to produce it. This problem has been compounded by the slow growth of programming productivity, especially for high-performance programs, over the past two decades. One way to bridge this gap is to make it possible for end users to develop programs in high-level domain-specific programming systems. In the past, a major impediment to the acceptance of such systems has been the poor performance of the resulting applications. To address this problem, we are developing a new compiler-based infrastructure, called TeleGen, that will make it practical to construct efficient domain-specific high-level languages from annotated component libraries. We call these languages telescoping languages, because they can be nested within one another. For programs written in telescoping languages, high performance and reasonable compilation times can be achieved by exhaustively analyzing the component libraries in advance to produce a language processor that recognizes and optimizes library operations as primitives in the language. The key to making this strategy practical is to keep compile times low by generating a custom compiler with extensive built-in knowledge of the underlying libraries. The goal is to achieve compile times that are linearly proportional to the size of the program presented by the user, rather than to the aggregate size of that program plus the base libraries.2001 Elsevier Science

show abstract

Section: Low-level Code Expansionmentioning

confidence: 77%

Telescoping Languages: A Strategy for Automatic Generation of Scientific Problem-Solving Systems from Annotated Libraries

Kennedy

Broom

Cooper

et al. 2001

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

show abstract

“…;;; for all x, (power x 10) = (power-10 x) = (power-10-linearized x) for compiling scientific code efficiently [2,3]. It is simple to modify Figure 2 to make it generate the let expressions, and again to simplify the residual code a bit in the base case (see appendix).…”

Section: An Effective Examplementioning

confidence: 99%

Partial Evaluation of the Euclidian Algorithm (Extended Version)

Danvy¹,

Goldberg²

1997

BRICS

View full text Add to dashboard Cite

Some programs are easily amenable to partial evaluation because their control flow clearly depends on one of their parameters. Specializing such programs with respect to this parameter eliminates the associated interpretive overhead. Some other programs, however, do not exhibit this interpreter-like behavior. Each of them presents a challenge for partial evaluation. The Euclidian algorithm is one of them, and in this article, we make it amenable to partial evaluation. We observe that the number of iterations in the Euclidian algorithm is bounded by a number that can be computed given either of the two arguments. We thus rephrase this algorithm using bounded recursion. The resulting program is better suited for automatic unfolding and thus for partial evaluation. Its specialization is efficient. Keywords: partial evaluation, scientific computation.

show abstract

“…An alternative, called online specialization, computes program parts as early as possible and takes decisions "on the fly" using only (and all) available information [Berlin and Weise 1990;Sahlin 1990;Weise et al 1991]. Figure 3 is a clear improvement over the unspecialized program, but can obviously be improved even more by "onthe-fly" reductions to a2͑n͒ ϭ if n ϭ 0 then 3 else a1͑a2͑n Ϫ 1͒͒ a1͑n͒ ϭ if n ϭ 0 then 2 else a1͑n Ϫ 1͒ ϩ 1 These methods often work better than offline methods, in particular on structured data that is partially static and …”

Section: Online and Offline Specializationmentioning

confidence: 99%

“…General programs for several diverse applications including orbit calculations (the n-body problem) and computations for electrical circuits have been speeded up by specialization to particular planetary systems and circuits [Baier et al 1994;Berlin and Weise 1990].…”

Section: Introduction To Partial Evaluationmentioning

confidence: 99%

An introduction to partial evaluation

Jones

1996

ACM Comput. Surv.

198

View full text Add to dashboard Cite

Partial evaluation provides a unifying paradigm for a broad spectrum of work in program optimization, compiling, interpretation and the generation of automatic program generators Ershov 1992;and Jones et al. 1993]. It is a program optimization technique, perhaps better called program specialization, closely related to but different from Jørring and Scherlis' staging transformations [1986]. It emphasizes, in comparison with Burstall and Darlington [1977] and Jørring and Scherlis [1986] and other program transformation work, full automation and the generation of program generators as well as transforming single programs. Much partial evaluation work to date has concerned automatic compiler generation from an interpretive definition of a programming language, but it also has important applications to scientific computing, logic programming, metaprogramming, and expert systems; some pointers are given later.

show abstract

Compiling scientific code using partial evaluation

Cited by 61 publications

References 3 publications

Telescoping Languages: A Strategy for Automatic Generation of Scientific Problem-Solving Systems from Annotated Libraries

Telescoping Languages: A Strategy for Automatic Generation of Scientific Problem-Solving Systems from Annotated Libraries

Partial Evaluation of the Euclidian Algorithm (Extended Version)

An introduction to partial evaluation

Contact Info

Product

Resources

About