Program locality analysis using reuse distance

Abstract

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 in general sequential code. These findings form the basis for a unified understanding of program Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies show this notice on the first page or initial screen of a display along with the full citation. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, to redistribute to lists, or to use any component of this work in other works requires prior specific permission and/or a fee. Permissions may be requested from Publications Dept., ACM, Inc., 2 Penn Plaza, Suite 701, New York, NY 10121-0701 USA, fax +1 (212) 869-0481, or permissions@acm.org. locality and its many facets. Concluding sections of the article present a taxonomy of related literature along five dimensions of locality and discuss the role of reuse distance in performance modeling, program optimization, cache and virtual memory management, and network traffic analysis.

Deconstructing iterative optimization

“…Zhong et al [2009] present two techniques to predict how program locality is affected across datasets; they consider 6 programs using up to 6 datasets each. Hsu et al [2002] compare the profiles generated using the 3 datasets of some SPECint2000 programs, and they observe significant program behavior variability across datasets for some programs.…”

Locality Principle Revisited: A Probability-Based Quantitative Approach

“…The histogram of reuse distances is used as a signature to quantify the temporal locality of a trace [20]. StatCache [3] provides an efficient way to estimate reuse distance histograms through sampling.…”

“…Between the two types of locality, the histogram of reuse distances [20] or LRU (least recently used) stack distances [15] computed from an address trace is used as a measure of temporal locality. For spatial locality, however, there is a lack of consensus for such a quantitative measure and several ad-hoc metrics [2][4] [11] [18] are proposed based on intuitive notions.…”

Coherent Profiles: Enabling Efficient Reuse Distance Analysis of Multicore Scaling for Loop-based Parallel Programs

“…RD analysis has also been used to analyze uniprocessor caches [1], [22], [29]. As discussed earlier, our work borrows reference groups from Zhong et al [1] to predict profile shift across core count scaling.…”