STM in the Small: Trading Generality for Performance in Software Transactional Memory

Aleksandar Dragojevic
2012
Article de conférence

Data structures implemented using software transactional memory (STM) have a reputation for being much slower than data structures implemented directly from low-level primitives such as atomic compare-and-swap (CAS). In this paper we present a specialized STM system (SpecTM) that allows the program to express additional knowledge about the particular operations being performed by transactions - e.g., using a separate API to write transactions that access small, fixed, numbers of memory locations. We show that data structures implemented using SpecTM offer essentially the same performance and scalability as implementations built directly from CAS. We present results using hash tables and skip lists on machines with up to 8 sockets and up to 128 hardware threads. Specialized transactions can be mixed with normal transactions, allowing fast-path operations to be specialized for greater performance, while allowing less common cases to be expressed using normal transactions for simplicity. We believe that SpecTM provides a “sweet spot” for expert programmers developing scalable data structures.

À propos de ce résultat

Cette page est générée automatiquement et peut contenir des informations qui ne sont pas correctes, complètes, à jour ou pertinentes par rapport à votre recherche. Il en va de même pour toutes les autres pages de ce site. Veillez à vérifier les informations auprès des sources officielles de l'EPFL.

Concepts associés

Chargement

Publications associées

Chargement

Aleksandar Dragojevic
2012
Article de conférence

Résumé

Official source

https://infoscience.epfl.ch/record/174888?ln=fr

À propos de ce résultat

Concepts associés (7)

Concepts associés

Chargement

Publications associées (3)

Publications associées

Chargement

Transactional memory

In computer science and engineering, transactional memory attempts to simplify concurrent programming by allowing a group of load and store instructions to execute in an atomic way. It is a concurrenc

Structure de données

En informatique, une structure de données est une manière d'organiser les données pour les traiter plus facilement. Une structure de données est une mise en œuvre concrète d'un type abstrait. Obj

Table de hachage

Une table de hachage est, en informatique, une structure de données qui permet une association clé–valeur, c'est-à-dire une implémentation du type abstrait tableau associatif. Son but principal est d

Log-Free Concurrent Data Structures

Tudor Alexandru David, Aleksandar Dragojevic, Rachid Guerraoui, Mihail Igor Zablotchi

Non-volatile RAM (NVRAM) makes it possible for data structures to tolerate transient failures, assuming however that programmers have designed these structures such that their consistency is preserved upon recovery. Previous ap- proaches are typically transactional and inherently make heavy use of logging, resulting in implementations that are significantly slower than their DRAM counterparts. In this paper, we introduce a set of techniques aimed at lock-free data structures that, in the large majority of cases, remove the need for logging (and costly durable store instructions) both in the data structure algorithm and in the associated memory management scheme. Together, these generic techniques enable us to design what we call log-free concurrent data structures, which, as we illustrate on linked lists, hash tables, skip lists, and BSTs, can provide several-fold performance improvements over previous transaction-based implementations, with overheads of the order of milliseconds for recovery after a failure. We also highlight how our techniques can be integrated into practical systems, by presenting a durable version of Memcached that maintains the performance of its volatile counterpart.

USENIX Association2018

Chargement

Log-Free Concurrent Data Structures

Tudor Alexandru David, Aleksandar Dragojevic, Rachid Guerraoui, Mihail Igor Zablotchi

Non-volatile RAM (NVRAM) makes it possible for data structures to tolerate transient failures, assuming however that programmers have designed these structures such that their consistency is preserved upon recovery. Previous approaches are typically transactional and inherently make heavy use of logging, resulting in implementations that are significantly slower than their DRAM counterparts. In this paper, we introduce a set of techniques aimed at lock-free data structures that, in the large majority of cases, remove the need for logging (and costly durable store instructions) both in the data structure algorithm and in the associated memory management scheme. Together, these generic techniques enable us to design what we call log-free concurrent data structures, which, as we illustrate on linked lists, hash tables, skip lists, and BSTs, can provide several-fold performance improvements over previous transaction-based implementations, with overheads of the order of milliseconds for recovery after a failure. We also highlight how our techniques can be integrated into practical systems, by presenting a durable version of Memcached that maintains the performance of its volatile counterpart.

USENIX ASSOC2018

Chargement

Optimistic Concurrency with OPTIK

Rachid Guerraoui, Vasileios Trigonakis

We introduce OPTIK, a new practical design pattern for designing and implementing fast and scalable concurrent data structures. OPTIK relies on the commonly-used technique of version numbers for detecting conflicting concurrent operations. We show how to implement the OPTIK pattern using the novel concept of OPTIK locks. These locks enable the use of version numbers for implementing very efficient optimistic concurrent data structures. Existing state-of-the-art lock-based data structures acquire the lock and then check for conflicts. In contrast, with OPTIK locks, we merge the lock acquisition with the detection of conflicting concurrency in a single atomic step, similarly to lock-free algorithms. We illustrate the power of our OPTIK pattern and its implementation by introducing four new algorithms and by optimizing four state-of-the-art algorithms for linked lists, skip lists, hash tables, and queues. Our results show that concurrent data structures built using OPTIK are more scalable than the state of the art.

ACM Press2016

Chargement