A feasibility study in iterative compilation

Kisuki, Toru; Knijnenburg, Peter M. W.; O’Boyle, Michael; Bodin, François; Wijshoff, Harry A. G.

doi:10.1007/bfb0094916

Cited by 47 publications

(51 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Earlier work in iterative compilation concentrated on finding good parameter settings for a few optimizations, such as loop unrolling and loop tiling [10,17,6]. In cases where exhaustive exploration was expensive, researchers used heuristic algorithms, such as grid-based searches, hill climbers, and genetic algorithms, to scan only a portion of the search space.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating Heuristic Optimization Phase Order Search Algorithms

Kulkarni

Whalley

Tyson

et al. 2007

International Symposium on Code Generation and Optimization (CGO'07)

View full text Add to dashboard Cite

show abstract

Section: Related Workmentioning

confidence: 99%

“…Prior research has found that no single ordering of phases can achieve optimal performance on all applications or functions [1,2,3,4,5,6]. The phase ordering problem is difficult since even after decades of research the relationships and interactions between optimization phases remain ill-understood.…”

Section: Introductionmentioning

confidence: 99%

Evaluating Heuristic Optimization Phase Order Search Algorithms

Kulkarni

Whalley

Tyson

et al. 2007

International Symposium on Code Generation and Optimization (CGO'07)

View full text Add to dashboard Cite

show abstract

“…To address this issue, a secondary goal of our work is to develop a general automatic tuning solution, which can be applied to many compilation techniques. In this regard, our work is related to tuning approaches that navigate a search space of possible transformation options, picking the best through empirical evaluation [5]. This method contrasts with work that directs the search process through performance models [6].…”

mentioning

confidence: 99%

Automatically Tuning Parallel and Parallelized Programs

Dave

Eigenmann

2010

Languages and Compilers for Parallel Computing

View full text Add to dashboard Cite

Abstract. In today's multicore era, parallelization of serial code is essential in order to exploit the architectures' performance potential. Parallelization, especially of legacy code, however, proves to be a challenge as manual efforts must either be directed towards algorithmic modifications or towards analysis of computationally intensive sections of code for the best possible parallel performance, both of which are difficult and time-consuming. Automatic parallelization uses sophisticated compiletime techniques in order to identify parallelism in serial programs, thus reducing the burden on the program developer. Similar sophistication is needed to improve the performance of hand-parallelized programs. A key difficulty is that optimizing compilers are generally unable to estimate the performance of an application or even a program section at compiletime, and so the task of performance improvement invariably rests with the developer. Automatic tuning uses static analysis and runtime performance metrics to determine the best possible compile-time approach for optimal application performance. This paper describes an offline tuning approach that uses a source-to-source parallelizing compiler, Cetus, and a tuning framework to tune parallel application performance. The implementation uses an existing, generic tuning algorithm called Combined Elimination to study the effect of serializing parallelizable loops based on measured whole program execution time, and provides a combination of parallel loops as an outcome that ensures to equal or improve performance of the original program. We evaluated our algorithm on a suite of hand-parallelized C benchmarks from the SPEC OMP2001 and NAS Parallel benchmarks and provide two sets of results. The first ignores hand-parallelized loops and only tunes application performance based on Cetus-parallelized loops. The second set of results considers the tuning of additional parallelism in hand-parallelized code. We show that our implementation always performs near-equal or better than serial code while tuning only Cetus-parallelized loops and equal to or better than hand-parallelized code while tuning additional parallelism.

show abstract

“…For example, GCC compilers include 38 options, roughly grouped into three optimization levels, O1 through O3. On the other hand, compiler optimizations interact in unpredictable manners, as many have observed [5,11,13,15,18]. A fast and effective orchestration algorithm to search for the best optimization combination for a program is desired.…”

Section: Introductionmentioning

confidence: 99%

“…ATLAS [20] starts with a parameterized, hand-coded set of matrix multiplication variants and evaluates them on the target machine to determine the optimum settings for that context. Similarly, Iterative Compilation [11] searches through the transformation space to find the best block sizes and unrolling factors. In more recent research [10], five different algorithms, genetic algorithm, simulated annealing, grid search, window search and random search are exploited to find the best blocking and unrolling parameters.…”

Section: Introductionmentioning

confidence: 99%