H. El-Rewini, T. G. Lewis, and H. H. Ali. Task scheduling in parallel and distributed systems. Prentice-Hall, Englewood Cli#s, NJ, USA, 1994.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....algorithm which maps the task graph on a set of processors. List scheduling is a general class of scheduling heuristics where tasks are assigned priorities and placed in a list in decreasing order of priority. Whenever tasks compete for processors, those with higher priorities are scheduled first [22]. The proposed algorithm accommodates pipelining and schedules tasks in a level based fashion starting with the entry tasks. It can be extended in a straightforward manner to handle multiple task graphs. The lower bound k min on the number of processors needed to schedule G i is given by its ....

H. El-Rewini, T. G. Lewis, and H. H. Ali, Task Scheduling in Parallel and Distributed Systems, Prentice Hall, Englewood Cliffs, NJ, 1994.

.... representation schemes [15 17] using alternative memory systems [18] or encouraging the maintenance of multiple species within a GA population [19 21] 3 Problem description We perform our experiments on a variable length GA applied to the problem of multiprocessor task scheduling [22]. The task scheduling problem begins with a task dependency graph which specifies the dependencies among a number of tasks that together compose a larger complete task. Figure 1 shows two example task dependency graphs. The goal of the GA is to assign tasks to a set of available parallel ....

El-Rewini, H., Lewis, T.G., Ali, H.H.: Task scheduling in parallel and distributed systems. Prentice Hall (1994)

....cost is the same for all three algorithms. Key Words: Loop scheduling, path driven scheduling, path generation, path mapping, generalized hypercube. 1 Introduction The problem of scheduling parallel program modules onto multiprocessor computers is known to be NP complete in general cases [5]. Many heuristic scheduling algorithms have been proposed in the literature [3] 5] 6] 8] 9] 10] 11] 12] 13] Some researchers introduced priority based algorithms, such as list [6] and free scheduling [8] Most of these are suitable for shared memory machines. When these algorithms ....

....scheduling, path generation, path mapping, generalized hypercube. 1 Introduction The problem of scheduling parallel program modules onto multiprocessor computers is known to be NP complete in general cases [5] Many heuristic scheduling algorithms have been proposed in the literature [3] [5], 6] 8] 9] 10] 11] 12] 13] Some researchers introduced priority based algorithms, such as list [6] and free scheduling [8] Most of these are suitable for shared memory machines. When these algorithms are applied to distributed memory machines, performance will degrade quickly ....

[Article contains additional citation context not shown here]

H. El-Rewini, T. G. Lewis and H. H. Ali, Task Scheduling in Parallel and Distributed Systems, Prentice Hall, 1994.

....is to achieve the minimum execution time of all the tasks satisfying the task precedence requirements. One of the approaches on this research area is the task processor mapping with weighted directed acyclic graphs (DAGs) which represent applications, and it has been known as NP Complete [2, 3, 6]. Although many scheduling heuristics have been proposed for a couple of decades, most of them are only devised for a homogeneous system. Recently a heterogeneous computing system is arousing the academic and commercial interests because of the good scalability and low building cost. For the high ....

....is more dicult than in case of a homogeneous system. In this study, we are interested in the static task scheduling for network of heterogeneous processors. Static scheduling is done at compile time with a directed acyclic graph (DAG) and the information of processors and a network architecture [2, 3]. Several heuristics for the heterogeneous environment have been proposed. The Dynamic Level Scheduling (DLS) algorithm is a greedy algorithm in that it selects a match of a task and a processor which gives the earliest start time [7] All of the ready tasks are the candidates for mapping on ....

El-Rewini, H., Lewis, T., Ali, H.: Task Scheduling in Parallel and Distributed Systems. Prentice Hall, Englewood Clis, New Jersey, 1994

....systems have to guarantee functional and timing correctness even in the presence of hardware and software failures. One way to guarantee that all timing and resource constraints will be met is to statically schedule all tasks to meet their deadlines. An off line scheduling strategy [3] [4] considers resource, precedence, and synchronization requirements of all tasks in the system and generates a feasible schedule that guarantees that the timing requirements of the tasks will be met. The scheduler is typically table driven [3] has low run time overhead, and can be easily verified ....

....task failures using temporal redundancy in non replicated processors enhances the safety and stability of the controlled process. We initially use an off line scheduling strategy to compute a feasible schedule for the distributed system. In this paper, the well known list scheduling (LS) strategy [4] is used to compute the schedule. This schedule is then used to generate contingency schedules for each task failure. Adaptive fault tolerance is provided in the contingency schedules by varying the QoS of the affected processor based on the spare capacity available. A fault tolerance layer is ....

[Article contains additional citation context not shown here]

H. El-Rewini et al., Task Scheduling in Parallel and Distributed Systems, Prentice Hall, Englewood Cliffs, NJ, 1994.

....an edge weighted directed acyclic graph (DAG) also called task graph or macro data ow graph, to a set of homogeneous processors to minimize the completion time. Since the static scheduling problem is NP complete in its general forms [6] and optimal solutions are known in restricted cases [3, 5, 7], there has been considerable research e ort in this area, resulting in many heuristic algorithms [19, 24, 4, 25, 20, 2, 14] In this paper, instead of suggesting a new scheduling algorithm, we present an algorithm that can improve the scheduling quality of the existing scheduling algorithms by ....

.... (MCP) algorithm is based on the as late as possible (ALAP) time of a node [24] The ALAP time is de ned as TL (n i ) T critical level(n i ) where T critical is the length of the critical path, and level(n i ) is the length of the longest path from node n i to an exit node, including node n i [5]. The MCP algorithm was designed to schedule a DAG on a bounded number of PEs. It sorts the node list in the increasing ALAP order. The rst node in the list is scheduled to the PE that allows the earliest start time, considering idle time slots. Then the node is deleted from the list and this ....

H. El-Rewini, T. G. Lewis, and H. H. Ali. Task Scheduling in Parallel and Distributed Systems. Prentice Hall, 1994.

No context found.

H. El-Rewini, T. G. Lewis, and H. H. Ali. Task scheduling in parallel and distributed systems. Prentice-Hall, Englewood Cli#s, NJ, USA, 1994.

No context found.

J. EL-REWINI, T. G. Lewis, and H. H. ALI. Task Scheduling in Parallel and Distributed Systems. Prentice-Hall, 1994.

No context found.

El-Rewini, H., Lewis, T.G., and Ali, H.H.: TASK SCHEDULING in PARALLEL and DISTRIBUTED SYSTEMS, PTR Prentice Hall (1994)

No context found.

H. El-Rewini, T. G. Lewis, and H. H. Ali. Task scheduling in parallel and distributed systems. Prentice Hall, 1994.

No context found.

H. El-Rewini, T. Lewis, and H. Ali. TASK SCHEDULING in PARALLEL and DISTRIBUTED SYSTEMS. PTR Prentice Hall, 1994.

No context found.

H. El-Rewini, H. Ali, T. Lewis, Task Scheduling in Parallel and Distributed Systems Prentice-Hall, 1994.

No context found.

El-Rewini, H., Lewis, T.G., and Ali, H.H., TASK SCHEDULING in PARALLEL and DISTRIBUTED SYSTEMS, pp.56--105, PTR Prentice Hall, 1994

No context found.

H. El-Rewini, T. G. Lewis, and H. H. Ali. Task scheduling in parallel and distributed systems. Prentice Hall, 1994.

No context found.

H. El-Rewini et al., Task Scheduling in Parallel and Distributed Systems, Prentice Hall, Englewood Cliffs, NJ, 1994.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC