We propose and evaluate a hybrid task scheduling method in order to reduce elapse time of parallel and distributed programs. Our task scheduling method combines the critical path (CP) scheduling method and the list scheduling method. The critical path scheduling method is efficient to assign tasks to processors, but it needs long time. The list scheduling methods is not efficient, but the scheduling time is very short. We have implemented a hybrid method which changes these scheduling methods by the number of tasks and workers. In the result, when we execute parallel and distributed programs based on 180 task graphs which include 100 tasks, faster than the original task scheduling methods at many cases.情報処理学会研究報告 高知県高知市(2006
International audienceWe study the problem of executing an application represented by a precedence t...
Efficient application scheduling is critical for achieving high performance in heterogeneous computi...
Task mapping and scheduling are two very difficult problems that must be addressed when a sequential...
This paper presents a hybrid scheduling methodology for task graphs to multiprocessor embedded syste...
International audienceIn this paper, we present an efffficient algorithm for compile time scheduling ...
Abstract — The purpose of multiprogramming is to maximize the CPU utilization through some process r...
Communicated by Susumu Matsumae This paper studies task scheduling algorithms which schedule a set o...
The evolution of computers is moving more and more towards multi-core processors and parallel progra...
AbstractExascale computing is fast becoming a mainstream research area. In order to realize exascale...
Abstract 1 In this paper, we survey algorithms that allocate a parallel program represented by an ed...
algorithm for compile-time static scheduling of task graphs onto multiprocessors is proposed. The pr...
In this paper, we propose a static scheduling algorithm for allocating task graphs to fully connecte...
Multiprocessor task scheduling plays a fundamental role in parallel applications and distributed net...
In this paper, we survey algorithms that allocate a parallel program represented by an edge-weighted...
Task Scheduling problem for heterogeneous systems is concerned with arranging the various tasks to b...
International audienceWe study the problem of executing an application represented by a precedence t...
Efficient application scheduling is critical for achieving high performance in heterogeneous computi...
Task mapping and scheduling are two very difficult problems that must be addressed when a sequential...
This paper presents a hybrid scheduling methodology for task graphs to multiprocessor embedded syste...
International audienceIn this paper, we present an efffficient algorithm for compile time scheduling ...
Abstract — The purpose of multiprogramming is to maximize the CPU utilization through some process r...
Communicated by Susumu Matsumae This paper studies task scheduling algorithms which schedule a set o...
The evolution of computers is moving more and more towards multi-core processors and parallel progra...
AbstractExascale computing is fast becoming a mainstream research area. In order to realize exascale...
Abstract 1 In this paper, we survey algorithms that allocate a parallel program represented by an ed...
algorithm for compile-time static scheduling of task graphs onto multiprocessors is proposed. The pr...
In this paper, we propose a static scheduling algorithm for allocating task graphs to fully connecte...
Multiprocessor task scheduling plays a fundamental role in parallel applications and distributed net...
In this paper, we survey algorithms that allocate a parallel program represented by an edge-weighted...
Task Scheduling problem for heterogeneous systems is concerned with arranging the various tasks to b...
International audienceWe study the problem of executing an application represented by a precedence t...
Efficient application scheduling is critical for achieving high performance in heterogeneous computi...
Task mapping and scheduling are two very difficult problems that must be addressed when a sequential...