Sista: Saving Optimized Code in Snapshots for Fast Start-Up

Abstract : Modern virtual machines for object-oriented languages such as Java HotSpot, Javascript V8 or Python PyPy reach high performance through just-in-time compilation techniques, involving on-the-fly optimization and deoptimization of the executed code. These techniques require a warm-up time for the virtual machine to collect information about the code it executes to be able to generate highly optimized code. This warm-up time required before reaching peak performance can be considerable and problematic. In this paper, we propose an approach, Sista (Speculative Inlining SmallTalk Architecture) to persist optimized code in a platform-independent representation as part of a snapshot. After explaining the overall approach, we show on a large set of benchmarks that the Sista virtual machine can reach peak performance almost immediately after start-up when using a snapshot where optimized code was persisted.
Type de document :
Communication dans un congrès
Proceedings of the 14th International Conference on Managed Languages and Runtimes, Sep 2017, Prague, Czech Republic. ACM, pp.1 - 11, 2017, 〈10.1145/3132190.3132201〉
Liste complète des métadonnées

Littérature citée [24 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01596321
Contributeur : Clement Bera <>
Soumis le : mercredi 27 septembre 2017 - 18:06:06
Dernière modification le : lundi 9 avril 2018 - 12:20:07
Document(s) archivé(s) le : jeudi 28 décembre 2017 - 14:09:15

Fichier

Hal.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Clément Béra, Eliot Miranda, Tim Felgentreff, Marcus Denker, Stéphane Ducasse. Sista: Saving Optimized Code in Snapshots for Fast Start-Up. Proceedings of the 14th International Conference on Managed Languages and Runtimes, Sep 2017, Prague, Czech Republic. ACM, pp.1 - 11, 2017, 〈10.1145/3132190.3132201〉. 〈hal-01596321〉

Partager

Métriques

Consultations de la notice

362

Téléchargements de fichiers

136