Tue 27 Oct 2015 16:18 - 16:42 at Grand Station 3 - Session 4, Empirical Studies

Modern JavaScript engines optimize hot functions using a JIT compiler along with type information gathered by an online profiler. However, the profiler’s information can be unsound and when unexpected types are encountered the engine must recover using an expensive mechanism called deoptimization. In this paper we describe a method to significantly reduce the number of deoptimizations observed by client-side JavaScript engines by using ahead-of-time profiling on the server-side. Unlike previous work on ahead-of-time profiling for statically-typed languages such as Java, our technique must operate on a dynamically-typed language, which significantly changes the required insights and methods to make the technique effective. We implement our proposed technique using the SpiderMonkey JavaScript engine, and we evaluate our implementation using three different kinds of benchmarks: the industry-standard Octane benchmark suite, a set of JavaScript physics engines, and a set of real-world websites from the Membench50 benchmark suite. We show that using ahead-of-time profiling provides significant performance benefits over the baseline vanilla SpiderMonkey engine.

Tue 27 Oct

dls2015
15:30 - 17:30: DLS - Session 4, Empirical Studies at Grand Station 3
dls201515:30 - 15:54
Talk
Beatrice ÅkerblomStockholm University, Tobias WrigstadUppsala University
dls201515:54 - 16:18
Talk
dls201516:18 - 16:42
Talk
Madhukar KedlayaUniversity of California, Santa Barbara, Behnam RobatmiliQualcomm Research, Ben HardekopfUC Santa Barbara
dls201516:42 - 17:06
Talk
Lars FischerUniversity of Duisburg-Essen, Essen, Germany, Stefan HanenbergUniversity of Duisburg-Essen
dls201517:06 - 17:30
Talk
Camille TeruelINRIA, Stéphane DucasseINRIA, France, Damien CassouLille 1 University, Marcus Denker INRIA Lille