Coz: Finding Code that Counts with Causal Profiling.
Improving performance is a central concern for software developers. To locate optimization opportunities, developers rely on software profilers. However, these profilers only report where programs spend their time: optimizing that code may have no impact on performance. Past profilers thus both wast...
| Published in: | Communications of the ACM Vol. 61; no. 6; pp. 91 - 100 |
|---|---|
| Main Authors: | , |
| Format: | Article |
| Published: |
Association for Computing Machinery
Jun2018
|
| Subjects: | |
| Online Access: | View this record in EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=hlh&AN=129988390&site=ehost-live header: @attributes: shortDbName: hlh uiTerm: 129988390 longDbName: Humanities International Complete uiTag: AN controlInfo: bkinfo: jinfo: jid: 00010782 ACM jtl: Communications of the ACM issn: 00010782 maglogo: N pubinfo: dt: Jun2018 vid: 61 iid: 6 pid: 68 pub: Association for Computing Machinery artinfo: ui: 129988390 10.1145/3205911 ppf: 91 ppct: 9 formats: tig: atl: Coz: Finding Code that Counts with Causal Profiling. aug: au: Curtsinger, Charlie Berger, Emery D. affil: Department of Computer Science, Grinnell College, USA College of Information and Computer Sciences, University of Massachusetts Amherst, USA su: Software engineering Dynamic programming Program transformation Computer programming software Computer programming management sug: subj: Software engineering Dynamic programming Program transformation Computer programming software Computer programming management ab: Improving performance is a central concern for software developers. To locate optimization opportunities, developers rely on software profilers. However, these profilers only report where programs spend their time: optimizing that code may have no impact on performance. Past profilers thus both waste developer time and make it difficult for them to uncover significant optimization opportunities. This paper introduces causal profiling. Unlike past profiling approaches, causal profiling indicates exactly where programmers should focus their optimization efforts, and quantifies their potential impact. Causal profiling works by running performance experiments during program execution. Each experiment calculates the impact of any potential optimization by virtually speeding up code: inserting pauses that slow down all other code running concurrently. The key insight is that this slowdown has the same relative effect as running that line faster, thus "virtually" speeding it up. We present Coz, a causal profiler, which we evaluate on a range of highly-tuned applications such as Memcached, SQLite, and the PARSEC benchmark suite. Coz identifies previously unknown optimization opportunities that are both significant and targeted. Guided by Coz, we improve the performance of Memcached by 9%, SQLite by 25%, and accelerate six PARSEC applications by as much as 68%; in most cases, these optimizations involve modifying under 10 lines of code. pubtype: Periodical doctype: Article src: R language: English refInfo: copyright: @attributes: flag: Y dt: @attributes: year: 2018 holdings: @attributes: islocal: N |
|---|