List scheduling with and without communication delays

Yang, Tao; Gerasoulis, Apostolos

doi:10.1016/0167-8191(93)90079-z

Cited by 138 publications

(67 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We said that Critical Path based policies seemed best suited, as shown by empirical tests with known communication delays [22], and also by bound results (see [11] and [10]). These empirical tests, and ours, show that the best priority rule is the following: for each task T i compute L * (i), the longest path that starts at T i , including processing times and communication delays, the processing time of the task without successor in the path but not the processing time of T i .…”

Section: Detailed Partially On-line Algorithmmentioning

confidence: 99%

“…4) and fixed communication delays, whereas RCP is optimal only if all tasks have the same duration (see [22]). However this is no longer true if the communication delays are allowed to vary.…”

Section: Join Graphsmentioning

confidence: 99%

“…One task T 1 is the immediate predecessor of all the others, which are final tasks. In the case of fixed communication delays, RCP or RCP * find the minimum makespan (see [22]). …”

Section: Fork Graphsmentioning

confidence: 99%

“…The priority of T i is proportional with L * (i). The heuristic that sequences ready tasks using the above priority is called RCP * in [22] (RCP if the processing time of T i is included).…”

Section: Detailed Partially On-line Algorithmmentioning

confidence: 99%

See 3 more Smart Citations

Parallel Machine Scheduling with Uncertain Communication Delays

2003

View full text Add to dashboard Cite

Abstract. This paper is concerned with scheduling when the data are not fully known before the execution. In that case computing a complete schedule off-line with estimated data may lead to poor performances. Some flexibility must be added to the scheduling process. We propose to start from a partial schedule and to postpone the complete scheduling until execution, thus introducing what we call a stabilization scheme. This is applied to the m machine problem with communication delays: in our model an estimation of the delay is known at compile time; but disturbances due to network contention, link failures, ... may occur at execution time. Hence the processor assignment and a partial sequencing on each processor are determined off-line. Some theoretical results for tree-like precedence constraints and an experimental study show the interest of this approach compared with fully on-line scheduling.

show abstract

Section: Detailed Partially On-line Algorithmmentioning

confidence: 99%

Section: Join Graphsmentioning

confidence: 99%

“…One task T 1 is the immediate predecessor of all the others, which are final tasks. In the case of fixed communication delays, RCP or RCP * find the minimum makespan (see [22]). …”

Section: Fork Graphsmentioning

confidence: 99%

Section: Detailed Partially On-line Algorithmmentioning

confidence: 99%

See 2 more Smart Citations

Parallel Machine Scheduling with Uncertain Communication Delays

2003

View full text Add to dashboard Cite

show abstract

“…The heuristic implementing this solution is a greedy list scheduling [22], called the Fault-Tolerance Based Active Replication strategy (FTBAR) algorithm. We present the scheduling algorithm in macrosteps, the superscript number in parentheses refers to the step of the heuristic, e.g.,…”

Section: Scheduling Heuristicmentioning

confidence: 99%