Dynamic scheduling of multi-processor tasks on clusters of clusters

Hunold, Sascha; Rauber, Thomas; Rünger, Gudula

doi:10.1109/clustr.2007.4629277

Cited by 5 publications

(5 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, RePA [22] is proposed as a dynamic scheduling algorithm to reduce the communication and redistribution costs by mapping child jobs to processors which are assigned to parent jobs (reuse processors). However, the algorithm makes an unrealistic assumption that jobs are always mapped to a single cluster.…”

Section: Related Workmentioning

confidence: 99%

Queue Waiting Time Aware Dynamic Workflow Scheduling in Multicluster Environments

余志峰

施巍松

2010

J. Comput. Sci. Technol.

View full text Add to dashboard Cite

Workflows are prevailing in scientific computation. Multicluster environments emerge and provide more resources, benefiting workflows but also challenging the traditional workflow scheduling heuristics. In a multicluster environment, each cluster has its own independent workload management system. Jobs are queued up before getting executed, they experience different resource availability and wait time if dispatched to different clusters. However, existing scheduling heuristics neither consider the queue wait time nor balance the performance gain with data movement cost. The proposed algorithm leverages the advancement of queue wait time prediction techniques and empirically studies if the tunability of resource requirements helps scheduling. The extensive experiment with both real workload traces and test bench shows that the queue wait time aware algorithm improves workflow performance by 3 to 10 times in terms of average makespan with relatively very low cost of data movement.

show abstract

Section: Related Workmentioning

confidence: 99%

Queue Waiting Time Aware Dynamic Workflow Scheduling in Multicluster Environments

余志峰

施巍松

2010

J. Comput. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…(Mixed-parallel applications are also referred to as "malleable tasks with precedence constraints" in the literature.) Prime candidates for mixed-parallel implementations are scientific workflows, as seen in existing workflow systems that support mixed parallelism [20,22,44]. For instance, mixed-parallel applications arise for instance in image processing applications that consist of a workflow of image filters, where some of these filters can be themselves implemented as data-parallel applications [18].…”

Section: Introductionmentioning

confidence: 99%

Scheduling mixed-parallel applications with advance reservations

Aida

Casanova

2009

Cluster Comput

View full text Add to dashboard Cite

This paper investigates the scheduling of mixedparallel applications, which exhibit both task and data parallelism, in advance reservations settings. Both the problem of minimizing application turn-around time and that of meeting a deadline are studied. For each several scheduling algorithms are proposed, some of which borrow ideas from previously published work in non-reservation settings. Algorithms are compared in simulation over a wide range of application and reservation scenarios. The main finding is that schedules computed using the previously published CPA algorithm can be adapted to advance reservation settings, notably resulting in low resource consumption and thus high efficiency.

show abstract

“…(Mixed-parallel applications are also referred to as "malleable tasks with precedence constraints" in the literature.) Prime candidates for mixedparallel implementations are scientific workflows, as seen in existing workflow systems that support mixed parallelism [45,27,25].…”

Section: Introductionmentioning

confidence: 99%

Scheduling mixed-parallel applications with advance reservations

Aida

Casanova

2008

Proceedings of the 17th International Symposium on High Performance Distributed Computing

View full text Add to dashboard Cite

This paper investigates the scheduling of mixed-parallel applications, which exhibit both task and data parallelism, in advance reservations settings. Both the problem of minimizing application turn-around time and that of meeting a deadline are studied. For each several scheduling algorithms are proposed, some of which borrow ideas from previously published work in non-reservation settings. Algorithms are compared in simulation over a wide range of application and reservation scenarios. The main finding is that schedules computed using the previously published CPA algorithm can be adapted to advance reservation settings, notably resulting in low resource consumption and thus high efficiency.

show abstract

Dynamic scheduling of multi-processor tasks on clusters of clusters

Cited by 5 publications

References 11 publications

Queue Waiting Time Aware Dynamic Workflow Scheduling in Multicluster Environments

Queue Waiting Time Aware Dynamic Workflow Scheduling in Multicluster Environments

Scheduling mixed-parallel applications with advance reservations

Scheduling mixed-parallel applications with advance reservations

Contact Info

Product

Resources

About