English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Locks have been widely used as an effective synchronization mechanism amongprocesses and threads. However, we observe that a large number of falseinter-thread dependencies (i.e., unnecessary lock contentions) exist during theprogram execution on multicore processors, thereby incurring significantperformance overhead. This paper presents a performance debugging framework,PERFPLAY, to facilitate a comprehensive and in-depth understanding of theperformance impact of unnecessary lock contentions. The core technique of ourdebugging framework is trace replay. Specifically, PERFPLAY records the programexecution trace, on the basis of which the unnecessary lock contentions can beidentified through trace analysis. We then propose a novel technique of tracetransformation to transform these identified unnecessary lock contentions inthe original trace into the correct pattern as a new trace free of unnecessarylock contentions. Through replaying both traces, PERFPLAY can quantify theperformance impact of unnecessary lock contentions. To demonstrate theeffectiveness of our debugging framework, we study five real-world programs andPARSEC benchmarks. Our experimental results demonstrate the significantperformance overhead of unnecessary lock contentions, and the effectiveness ofPERFPLAY in identifying the performance critical unnecessary lock contentionsin real applications.

On Performance Debugging of Unnecessary Lock Contentions on Multicore Processors: A Replay-based Approach