Journal ArticleDOI
Coordinating Programs in the Network of Tasks Model
Reads0
Chats0
TLDR
The Network of Tasks (NOT) model allows adaptive node programs written in a variety of parallel languages to be connected together in an almost acyclic task graph to be regarded as a coordination language that is transparent about performance.Abstract:
The Network of Tasks (NOT) model allows adaptive node programs written in a variety of parallel languages to be connected together in an almost acyclic task graph. The main difference between NOT and other task graphs is that it is designed to make the performance of the graph predictable from knowledge of the performance of the component node programs and the visible structure of the graph. It can therefore be regarded as a coordination language that is transparent about performance. For large-scale computations that are distributed to geographically-distributed compute servers, the NOT model helps programmers to plan, assemble, schedule, and distribute their problems.read more
Citations
More filters
Proceedings Article
Towards an Architecture-Independent Analysis of Parallel Algorithms (Extended Abstract)
TL;DR: It would be very interesting if the authors could combine stages (3) and (4) into a single step whereby the performance of the algorithm is measured as the makespan of the schedule (elapsed time for computing the last result).
Book ChapterDOI
On compiler support for mixed task and data parallelism
TL;DR: This chapter discusses implementation issues of the TwoL model as an open framework and its internal algorithms and data structures and fast parallel matrix multiplication algorithms are discussed to illustrate the applicability of approach.
Journal ArticleDOI
Coordination as a Paradigm for Systems Integration
TL;DR: A coordination framework comprises three fundamental elements: separation of concerns between what is being computed and how the entities that comprise a system of computational units forming an application interact with each other.
Proceedings ArticleDOI
Semi-dynamic Scheduling of Parallel Tasks for Heterogeneous Clusters
TL;DR: Experimental results for several scientific applications show that the semi-dynamic execution scheme leads to lower execution times compared to a static execution on a tightly coupled heterogeneous platform.
References
More filters
Book
Programming from specifications
TL;DR: This second edition features substantial restructuring of earlier material, streamlining the introduction of programming language features; simplified presentation of procedures, parameters and recursion; an expanded chapter on data refinement, giving the much simpler laws that specialize to functional abstractions.
Journal ArticleDOI
The Parallel Evaluation of General Arithmetic Expressions
TL;DR: It is shown that arithmetic expressions with n ≥ 1 variables and constants; operations of addition, multiplication, and division; and any depth of parenthesis nesting can be evaluated in time 4 log 2 + 10(n - 1) using processors which can independently perform arithmetic operations in unit time.
Proceedings ArticleDOI
The Globus project: a status report
Ian Foster,Carl Kesselman +1 more
TL;DR: The progress that has been achieved to date in the development of the Globus toolkit, a set of core services for constructing grid tools and applications, is described and the GUSTO testbed, constructed to enable large-scale evaluation of Globus technologies, is discussed.
Journal ArticleDOI
Towards an architecture-independent analysis of parallel algorithms
TL;DR: A simple and efficient method for evaluating the performance of an algorithm, rendered as a directed acyclic graph, on any parallel computer is presented and its application to several common algorithms shows that it is surprisingly accurate.
Journal ArticleDOI
Questions and Answers about BSP
TL;DR: This paper answers the most common questions asked about BSP and justifies its claim to be a major step forward in parallel programming.
Related Papers (5)
Tlib-a library to support programming with hierarchical multi-processor tasks
Thomas Rauber,Gudula Rünger +1 more
A transformation approach to derive efficient parallel implementations
Thomas Rauber,Gudula Rünger +1 more
A new model for integrated nested task and data parallel programming
Jaspal Subhlok,Bwolen Yang +1 more