Pthreads for Dynamic and Irregular Parallelism

Narlikar, Girija; Blelloch, Guy E.

doi:10.1109/sc.1998.10005

Cited by 11 publications

(7 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The results indicate that the new scheduler, unlike the existing FIFO scheduler, allows irregular and dynamic parallel programs written with a very large number of Pthreads to execute efficiently [Narlikar and Blelloch 1998]. We are currently working on methods to further improve the scheduling algorithm, particularly to provide better support for fine-grained threads.…”

Section: Summary and Discussionmentioning

confidence: 99%

Space-efficient scheduling of nested parallelism

Narlikar

Blelloch

1999

ACM Trans. Program. Lang. Syst.

View full text Add to dashboard Cite

Many of today's high-level parallel languages support dynamic, fine-grained parallelism. These languages allow the user to expose all the parallelism in the program, which is typically of a much higher degree than the number of processors. Hence an efficient scheduling algorithm is required to assign computations to processors at runtime. Besides having low overheads and good load balancing, it is important for the scheduling algorithm to minimize the space usage of the parallel program. This article presents an on-line scheduling algorithm that is provably space efficient and time efficient for nested-parallel languages. For a computation with depth D and serial space requirement S 1 , the algorithm generates a schedule that requires at most S 1 + O(K · D · p) space (including scheduler space) on p processors. Here, K is a user-adjustable runtime parameter specifying the net amount of memory that a thread may allocate before it is preempted by the scheduler. Adjusting the value of K provides a trade-off between the running time and the memory requirement of a parallel computation. To allow the scheduler to scale with the number of processors, we also parallelize the scheduler and analyze the space and time bounds of the computation to include scheduling costs. In addition to showing that the scheduling algorithm is space and time efficient in theory, we demonstrate that it is effective in practice. We have implemented a runtime system that uses our algorithm to schedule lightweight parallel threads. The results of executing parallel programs on this system show that our scheduling algorithm significantly reduces memory usage compared to previous techniques, without compromising performance.

show abstract

Section: Summary and Discussionmentioning

confidence: 99%

Space-efficient scheduling of nested parallelism

Narlikar

Blelloch

1999

ACM Trans. Program. Lang. Syst.

View full text Add to dashboard Cite

show abstract

“…Recent work has resulted in run-time scheduling techniques that minimize completion time and memory use of the generated threads [11,8,27]. The automatic construction of threads of appropriate granularity is currently being investigated by several researchers [26,19].…”

Section: The Thread-based Approachmentioning

confidence: 99%

Irregular Computations in Fortran – Expression and Implementation Strategies

Prins

Chatterjee

Simons

1999

Scientific Programming

View full text Add to dashboard Cite

Modern dialects of Fortran enjoy wide use and good support on high-performance computers as performance-oriented programming languages. By providing the ability to express nested data parallelism, modern Fortran dialects enable irregular computations to be incorporated into existing applications with minimal rewriting and without sacrificing performance within the regular portions of the application. Since performance of nested data-parallel computation is unpredictable and often poor using current compilers, we investigate threading and flattening, two source-to-source transformation techniques that can improve performance and performance stability. For experimental validation of these techniques, we explore nested data-parallel implementations of the sparse matrix-vector product and the Barnes-Hut n-body algorithm by hand-coding thread-based (using OpenMP directives) and flattening-based versions of these algorithms and evaluating their performance on an SGI Origin 2000 and an NEC SX-4, two shared-memory machines.

show abstract

“…Unter diesen Voraussetzungen machen sich diese Probleme noch nicht dominierend bemerkbar, so daß z.B. für N-Körper-Algorithmen auf einer Sun Enterprise 5000 SMP mit acht Prozessoren fast optimale Speedups erreicht wurden [30]. Für Maschinen mit verteiltem Speicher und vielen Prozessoren dürfte eine Umsetzung mit guten Speedups deutlich schwieriger sein.…”

Section: Kontrollfädenunclassified

Verschachtelter Datenparallelismus – Ein einführender Überblick

Pfannenstiel

1999

Informatik Forsch Entw

View full text Add to dashboard Cite

Zusammenfassung. Datenparallele Programmiermodelle sind derzeit die erfolgreichsten Programmiermodelle für Parallelrechner, sowohl was die Effizienz der Ausführung als auch die Komplexität der Programmierung angeht. Es ist bisher jedoch nicht gelungen, ein paralleles Programmiermodell zu entwickeln, das sowohl konzeptionell einfach und abstrakt ist, als auch effizient auf die Fülle paralleler Rechnerarchitekturen zuübertragen ist. Das verschachtelt datenparallele Programmiermodell besitzt einige der gewünschten Eigenschaften eines parallelen Programmiermodells. Im Gegensatz zu den gängigen flach datenparallelen Modellen gibt es in verschachtelt datenparallelen Modellen Mechanismen, die irreguläre Datenstrukturen und Berechnungen direkt unterstützen. Dieser Artikel stellt einen kollektionsbasierten Ansatz für verschachtelten Datenparallelismus vor, gibt einenÜberblicküber den Stand der Forschung und zeigt offene Fragen in diesem Gebiet auf.Abstract. Today, data-parallel programming models are the most successful programming models for parallel computers both in terms of efficiency of execution and ease of use for the programmer. However, there is no parallel programming model that is conceptually simple and abstract, and that can be ported efficiently to the variety of parallel architectures available. The nested data-parallel programming model has some of the desired properties of a parallel programming model. In contrast to flat data parallel models, with this model it is possible to express irregular data structures and irregular parallel computations directly. In this paper, a collection-oriented approach to nested data parallelism is introduced. The state of the art of related research is presented and open questions are identified.

show abstract

Pthreads for Dynamic and Irregular Parallelism

Cited by 11 publications

References 33 publications

Space-efficient scheduling of nested parallelism

Space-efficient scheduling of nested parallelism

Irregular Computations in Fortran – Expression and Implementation Strategies

Verschachtelter Datenparallelismus – Ein einführender Überblick

Contact Info

Product

Resources

About