Skip to main content

Efficient and Thread-Safe Objects for Dynamically-Typed Languages

Daloze, Benoit, Marr, Stefan, Bonetta, Daniele, Mössenböck, Hanspeter (2016) Efficient and Thread-Safe Objects for Dynamically-Typed Languages. ACM SIGPLAN Notices, 51 (10). pp. 642-659. ISSN 0362-1340. E-ISSN 1558-1160. (doi:10.1145/2983990.2984001)

PDF - Author's Accepted Manuscript
Download (427kB) Preview
[img]
Preview
Official URL
http://doi.org/10.1145/2983990.2984001

Abstract

We are in the multi-core era. Dynamically-typed languages are in widespread use, but their support for multithreading still lags behind. One of the reasons is that the sophisticated techniques they use to efficiently represent their dynamic object models are often unsafe in multithreaded environments. This paper defines safety requirements for dynamic object models in multithreaded environments. Based on these requirements, a language-agnostic and thread-safe object model is designed that maintains the efficiency of sequential approaches. This is achieved by ensuring that field reads do not require synchronization and field updates only need to synchronize on objects shared between threads. Basing our work on JRuby+Truffle, we show that our safe object model has zero overhead on peak performance for thread-local objects and only 3% average overhead on parallel benchmarks where field updates require synchronization. Thus, it can be a foundation for safe and efficient multithreaded VMs for a wide range of dynamic languages.

Item Type: Article
DOI/Identification number: 10.1145/2983990.2984001
Uncontrolled keywords: Dynamically-sized objects, Dynamic languages, Concurrency, Virtual Machine, Java, Truffle, Graal, Ruby
Divisions: Faculties > Sciences > School of Computing
Depositing User: Stefan Marr
Date Deposited: 06 Oct 2017 12:48 UTC
Last Modified: 05 Jul 2019 14:50 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/63818 (The current URI for this page, for reference purposes)
Marr, Stefan: https://orcid.org/0000-0001-9059-5180
  • Depositors only (login required):

Downloads

Downloads per month over past year