This paper presents a general framework for determining average program execution times and their variance, based on the program's interval structure and control dependence graph. Average execution times and variance values are computed using frequency information from an optimized counter-based execution profile of the program. 1 Introduction It is important for a compiler to obtain estimates of execution times for subcomputations of an input program, if it is to attempt optimizations related to overhead values in the target architecture. In earlier work [SH86a, SH86b, Sar87, Sar89], we used estimates of execution times to facilitate the automatic partitioning and scheduling of programs written in the singleassignment language, Sisal, for parallel execution on multiprocessors. In this paper, we present a general framework for estimating average execution times in a program. This approach is based on the interval structure [ASU86] and the control dependence relation [FOW87], both of w...

Profiling can accurately analyze program behavior for select data inputs. We show that profiling can also predict program locality for inputs other than profiled ones. Here locality is defined by the distance of data reuse. Studying whole-program data reuse may reveal global patterns not apparent in short-distance reuses or local control flow. However, the analysis must meet two requirements to be useful. The first is efficiency. It needs to analyze all accesses to all data elements in full-size benchmarks and to measure distance of any length and in any required precision. The second is predication. Based on a few training runs, it needs to classify patterns as regular and irregular and, for regular ones, it should predict their (changing) behavior for other inputs. In this paper, we show that these goals are attainable through three techniques: approximate analysis of reuse distance (originally called LRU stack distance), pattern recognition, and distance-based sampling. When tested on 15 integer and floating-point programs from SPEC and other benchmark suites, our techniques predict with on average 94% accuracy for data inputs up to hundreds times larger than the training inputs. Based on these results, the paper discusses possible uses of this analysis.

On modern computer systems, the memory performance of an application depends on its locality. For a single execution, locality-correlated measures like average miss rate or working-set size have long been analyzed using reuse distance—the number of distinct locations accessed between consecutive accesses to a given location. This article addresses the analysis problem at the program level, where the size of data and the locality of execution may change significantly depending on the input. The article presents two techniques that predict how the locality of a program changes with its input. The first is approximate reuse-distance measurement, which is asymptotically faster than exact methods while providing a guaranteed precision. The second is statistical prediction of locality in all executions of a program based on the analysis of a few executions. The prediction process has three steps: dividing data accesses into groups, finding the access patterns in each group, and building parameterized models. The resulting prediction may be used on-line with the help of distance-based sampling. When evaluated on fifteen benchmark applications, the new techniques predicted program locality with good accuracy, even for test executions that are orders of magnitude larger than the training executions. The two techniques are among the first to enable quantitative analysis of whole-program locality and

To improve performance, data reorganization needs locality models to identify groups of data that have reference affinity. Much past work is based on access frequency and does not consider accessing time directly. In this paper, we propose a new model of reference affinity. This model considers the distance between data accesses in addition to the frequency. Affinity groups defined by this model are consistent and have a hierarchical structure. The former property ensures the profitability of data packing, while the latter supports data packing for storage units of different sizes. We then present a statistical clustering method that identifies affinity groups among structure fields and data arrays by analyzing training runs of a program. When used by structure splitting and array regrouping, the new method improves the performance of two test programs by up to 31%. The new data layout is significantly better than that produced by the programmer or by static compiler analysis.

absty.act tntez,pretation of orograrns is a mathematical r'rodel for static analysis of prograns: a ne\ { interpretation is given to the Drogran text and this a1lows the building of a system of esuat ions. The anal ysi s of the program tl"Ien consists in verifying thet a solution orovided by the user is correct, or in computing this solution, or else in discoverin3 a good annroxirnation of this solution. * Attach6 de Recherche au C.N.R.S". Laboratoire Associ6 no 7. * * This work was supported by IRIA-SESORI under grants 75-035 and 76-160.? The abstract interpret"ation of orograms \ras fornalized in Cousot I 1977a1. 0n1y a few lattice theoretical results are recalled here for the oaper to be self-contained. Our main purpose is to shorv that most Droqram analysis technicrues may be understood and nodelled as abstract interpretations of prograns. This is illustrated by the following examoles:- Global data flow analysis,