Multicore Scheduling for Lightweight Communicating Processes

Ritson, Carl G.; Sampson, Adam T.; Barnes, Frederick R.M.

doi:10.1007/978-3-642-02053-7_9

Cited by 12 publications

(4 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As well as simplification in application logic, safe and efficient exploitation of multicore processors follows automatically. At least, this is true for the occam-π multiprocessing language [Barnes and Welch 2004;Welch and Barnes 2005a, 2005bRitson et al 2009;Sampson et al 2010a; Barnes et al 2010a], used in this article, and the JCSP [Welch 2000;Welch et al 2007;Welch and Austin 2010] and C++CSP [Brown and Welch 2003;Brown 2007] libraries for Java and C++.…”

Section: Concurrency Is Our Friendmentioning

confidence: 90%

Santa Claus

Welch

Pedersen

2010

ACM Trans. Program. Lang. Syst.

View full text Add to dashboard Cite

With the commercial development of multicore processors, the challenges of writing multithreaded programs to take advantage of these new hardware architectures are becoming more and more pertinent. Concurrent programming is necessary to achieve the performance that the hardware offers. Traditional approaches present concurrency as an advanced topic: they have proven difficult to use, reason about with confidence, and scale up to high levels of concurrency. This article reviews process-oriented design, based on Hoare's algebra of Communicating Sequential Processes (CSP), and proposes that this approach to concurrency leads to solutions that are manageable by novice programmers; that is, they are easy to design and maintain, that they are scalable for complexity, obviously correct, and relatively easy to verify using formal reasoning and/or model checkers. These solutions can be developed in conventional programming languages (through CSP libraries) or specialized ones (such as occam-π) in a manner that directly reflects their formal expression. Systems can be developed without needing specialist knowledge of the CSP formalism, since the supporting mathematics is burnt into the tools and languages supporting it. We illustrate these concepts with the Santa Claus problem, which has been used as a challenge for concurrency mechanisms since 1994. We consider this problem as an example control system, producing external signals reporting changes of internal state (that model the external world). We claim our occam-π solution is correct-by-design, but follow this up with formal verification (using the FDR model checker for CSP) that the system is free from deadlock and livelock, that the produced control signals obey crucial ordering constraints, and that the system has key liveness properties.

show abstract

Section: Concurrency Is Our Friendmentioning

confidence: 90%

Santa Claus

Welch

Pedersen

2010

ACM Trans. Program. Lang. Syst.

View full text Add to dashboard Cite

show abstract

“…In contrast to Molecule, it does not feature resource control, message poisoning or higher-order channel interfaces. Its CCSP scheduler [Ritson et al 2009] exploits dynamically application-level communication patterns like Molecule's flow parallel scheduler. The design of the flow parallel scheduler is more modular and simpler: it is orthogonal to batching and delegates parallel task execution to an external thread pool, e.g.…”

Section: Related Workmentioning

confidence: 99%

“…The execution of parallel process networks is known to suffer inherently from a massive amount of context switches [Ritson et al 2009]. To alleviate this issue, we devised a scheduler that exploits our high-level abstractions to eliminate expensive and unnecessary kernel-level context switches dynamically, whenever the data flow is purely sequential.…”

Section: Introductionmentioning

confidence: 99%

Molecule

Bocq¹,

Daenen²

2012

Proceedings of the ACM International Conference on Object Oriented Programming Systems Languages and Applications

View full text Add to dashboard Cite

Molecule is a domain specific language library embedded in Scala for easing the creation of scalable and modular concurrent applications on the JVM. Concurrent applications are modeled as parallel process networks that exchange information over mobile and type-safe messaging interfaces.In this paper, we present a concurrent programming environment that combines functional and imperative programming. Using a monad, we structure the sequential or parallel coordination of user-level threads, without JVM modifications or compiler support. Our mobile channel interfaces expose reusable and parallelizable higher-order functions, as if they were streams in a lazily evaluated functional programming language. The support for graceful termination of entire process networks is simplified by integrating channel poisoning with monadic exceptions and resource control. Our runtime and system-level interfaces leverage message batching and a novel flow parallel scheduler to limit expensive context switches in multicore environments. We illustrate the expressiveness and performance benefits on a 24core AMD Opteron machine with three classical examples: a thread ring, a genuine prime sieve and a chameneos-redux.

show abstract

“…occam-π's excellent support for safe, lightweight, message-passing concurrency and its highly-efficient multicore runtime system make it straightforward to build agent-based simulations in which agents are implemented directly as lightweight processes. Such process-oriented simulations automatically make good use of modern multicore processors, and can be distributed over clusters of machines [5,6]. However, occam-π's limited set of data types and lack of bindings for external libraries makes data analysis and visualisation more complex than in other languages.…”

Section: Prototype Implementationmentioning

confidence: 99%

The best of most worlds

Sampson

Andrews

2010

Proceedings of the 9th Workshop on Parallel/High-Performance Object-Oriented Scientific Computing

Self Cite

View full text Add to dashboard Cite

Computing techniques are increasingly being used in scientific research to tackle a diverse set of problems. An example is complex systems research, which focuses on the use of computer simulations to explore, understand and describe the real-world system under study. These simulations are often sophisticated pieces of software with numerous design trade-offs between performance and ease of development and use. We propose a simulation framework for complex systems simulation that allows each component of a simulation-for example visualisation, or data analysis-to be developed in the most appropriate language. The framework uses the concept of shared objects to communicate data between simulation components. We present here a detailed motivation for multilingual simulations, an outline design and prototype for the simulation framework, and discuss future plans for the framework.

show abstract

Multicore Scheduling for Lightweight Communicating Processes

Cited by 12 publications

References 36 publications

Santa Claus

Santa Claus

Molecule

The best of most worlds

Contact Info

Product

Resources

About