The reflexive CHAM and the join-calculus

Fournet, Cédric; Gonthier, Georges

doi:10.1145/237721.237805

Cited by 341 publications

(332 citation statements)

References 19 publications

Supporting

Mentioning

328

Contrasting

Unclassified

Order By: Relevance

“…Join methods [17] are defined by a set of method fragments and their body is only executed when all method fragments are invoked. Implementation of Join methods can be found in C# [18] and Join Java [19].…”

Section: Related Workmentioning

confidence: 99%

Parallel actor monitors: Disentangling task-level parallelism from data partitioning in the actor model

Scholliers

Tanter

Meuter

2014

Science of Computer Programming

View full text Add to dashboard Cite

While the actor model of concurrency is well appreciated for its ease of use, its scalability is often criticized. Indeed, the fact that execution within an actor is sequential prevents certain actor systems to take advantage of multicore architectures. In order to combine scalability and ease of use, we propose Parallel Actor Monitors (PAM), as a means to relax the sequentiality of intra-actor activity in a structured and controlled way. A PAM is a modular, reusable scheduler that permits to introduce intra-actor parallelism in a local and abstract manner. PAM allows the stepwise refinement of local parallelism within a system on a per-actor basis, without having to deal with low-level synchronization details and locks. We present the general model of PAM and its instantiation in the AmbientTalk language. Benchmarks confirm the expected performance gain.

show abstract

Section: Related Workmentioning

confidence: 99%

Parallel actor monitors: Disentangling task-level parallelism from data partitioning in the actor model

Scholliers

Tanter

Meuter

2014

Science of Computer Programming

View full text Add to dashboard Cite

show abstract

“…We shall focus on a monadic version of the join calculus [5,7], writing its operational semantics in terms of a reduction system as in [15]. For a thorough introduction the reader is referred to the literature.…”

Section: The Join Calculusmentioning

confidence: 99%

“…The join calculus [5,7] is an algebra of mobile processes with asynchronous namepassing communication that simplifies the π-calculus by enforcing the hypothesis of unique receptors. This means that there is at most one process receiving messages on a name, and is a distinctive feature of the join calculus that makes it suitable for distributed implementations, as channels may be allocated at their receptor process.…”

Section: Introductionmentioning

confidence: 99%

High-Level Petri Nets as Type Theories in the Join Calculus

Buscemi

Sassone

2001

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. We study the expressiveness of the join calculus by comparison with (generalised, coloured) Petri nets and using tools from type theory. More precisely, we consider four classes of nets of increasing expressiveness, Π i, introduce a hierarchy of type systems of decreasing strictness, ∆i, i = 0, . . . , 3, and we prove that a join process is typeable according to ∆i if and only if it is (strictly equivalent to) a net of class Πi. In the details, Π0 and Π1 contain, resp., usual place/transition and coloured Petri nets, while Π2 and Π3 propose two natural notions of high-level net accounting for dynamic reconfiguration and process creation and called reconfigurable and dynamic Petri nets, respectively.

show abstract

“…If a channel end-point is seen passing through, the firewall must decide whether to allow communication on that channel, and if so it must create a forwarder for it. So, a channel through a firewall must really be handled as two channels connected by a filter [20].…”

Section: Dynamic Connectivitymentioning

confidence: 99%

“…Asynchronous messages simplify the requirements for distributed synchronization [32], but they still do not localize the management of communication decisions. The join-calculus [20] approaches this problem by rooting each channel at a particular process; this provides a single place where synchronization is resolved. LLinda [18] is a formalization of Linda [15] using process calculi techniques; as in distributed versions of Linda, LLinda has multiple distributed tuple spaces, each with its local synchronization manager.…”

Section: Distributionmentioning

confidence: 99%