Input-Sensitive Profiling

Abstract

In this article we present a building block technique and a toolkit towards automatic discovery of workload-dependent performance bottlenecks. From one or more runs of a program, our profiler automatically measures how the performance of individual routines scales as a function of the input size, yielding clues to their growth rate. The output of the profiler is, for each executed routine of the program, a set of tuples that aggregate performance costs by input size. The collected profiles can be used to produce performance plots and derive trend functions by statistical curve fitting techniques. A key feature of our method is the ability to automatically measure the size of the input given to a generic code fragment: to this aim, we propose an effective metric for estimating the input size of a routine and show how to compute it efficiently. We discuss several examples, showing that our approach can reveal asymptotic bottlenecks that other profilers may fail to detect and can provide useful characterizations of the workload and behavior of individual routines in the context of mainstream applications, yielding several code optimizations as well as algorithmic improvements. To prove the feasibility of our techniques, we implemented a Valgrind tool called aprof and performed an extensive experimental evaluation on the SPEC CPU2006 benchmarks. Our experiments show that aprof delivers comparable performance to other prominent Valgrind tools, and can generate informative plots even from single runs on typical workloads for most algorithmically-critical routines

Improving Program Comprehension through Dynamic Code Analysis

“…Aprof is a Valgrind tool designed to help developers identify inefficiencies in code [5]. It is input-sensitive, that means that on top of call graph, it takes into account the input for methods and measures their performance based on the workload received.…”

Context-sensitive delta inference for identifying workload-dependent performance bottlenecks

“…Existing approaches [16,19] that infer complexity models using profiles of multiple workloads are context-insensitive. To show effectiveness of context-sensitive analysis and answer RQ3, we compare the number of complexity transitions identified by using context-sensitive analysis and contextinsensitive analysis.…”

“…Goldsmith et al [19] propose an approach that fits performance measurements of clusters of basic blocks to workload sizes. Zaparanuks et al [38] propose an approach that infers an empirical cost function of an application automatically, and Coppa et al [16] propose an approach that measures the size of the input given to a generic code fragment. Unlike their model inference based on sorted or random inputs, our approach iteratively refines the inferred models based on the model accuracy from the previous iteration.…”

“…To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. ISSTA '13, July [15][16][17][18][19][20]2013, Lugano, Switzerland Copyright 13 ACM 978-1-4503-2159-4/13/07 ...$10.00.…”

Program performance spectrum

“…Recent works [10,24] have proposed to extend the traditional profiling technique by determining an empirical cost function. Such a cost function is found automatically and it captures an approximate cost of the program with respect to different inputs (and in particular, with respect to different input sizes).…”