Scheduling strategies for mapping application workflows onto the grid

Mandal, Anirban; Kennedy, Ken; Koelbel, Charles; Marin, Gabriel; Mellor-Crummey, John; Liu, B.; Johnsson, Lennart

doi:10.1109/hpdc.2005.1520947

Cited by 132 publications

(76 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Workflows application scheduling on grids [16], [26], [27] is an active area of research. Workflow scheduling has largely focused on heuristic techniques using performance models to qualitatively select resources and map tasks to the resources that have good performance [12].…”

Section: Related Workmentioning

confidence: 99%

“…For example, the study in [10] shows that the mean time between failures (MTBF) on Grid5000 [6] is only around 12 minutes. Hence, not only is managing and scheduling workflow applications a hard problem and studied in detail [16], [27], [26], additional challenges in providing reliability to workflow executions arise because of the unreliable nature of the underlying hardware and software.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Combined Fault Tolerance and Scheduling Techniques for Workflow Applications on Computational Grids

Yang

Mandal

Koelbel

et al. 2009

2009 9th IEEE/ACM International Symposium on Cluster Computing and the Grid

View full text Add to dashboard Cite

More and more complex scientific workflows are now executed on computational grids. In addition to the challenges of managing and scheduling these workflows, additional reliability challenges arise because of the unreliable nature of large-scale grid infrastructure. Fault tolerance mechanisms like over-provisioning and checkpoint-recovery are used in current grid application management systems to address these reliability challenges. In this work, we propose new approaches that combine these fault tolerance techniques with existing workflow scheduling algorithms. We present a study on the effectiveness of the combined approaches by analyzing their impact on the reliability of workflow execution, workflow performance and resource usage under different reliability models, failure prediction accuracies and workflow application types.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Combined Fault Tolerance and Scheduling Techniques for Workflow Applications on Computational Grids

Yang

Mandal

Koelbel

et al. 2009

2009 9th IEEE/ACM International Symposium on Cluster Computing and the Grid

View full text Add to dashboard Cite

show abstract

“…Furthermore, our experiments demonstrate that inaccurate performance models lead to bad schedules [9]. The goal of the VGrADS Project is to make Grid programming easier rather than more difficult, so requiring that the developer construct performance models by hand is out of the question.…”

Section: Schedulingmentioning

confidence: 89%

“…Finally, as we will see in Section 4.2, it makes it possible to incorporate estimated batch queue waiting times into the schedule as extra delays between job steps. Our experiments have shown that off-line scheduling can produce dramatically better workflow completion times, in many cases by factors greater than 2, than dynamic approaches [2,9].…”

Section: Schedulingmentioning

confidence: 90%

See 1 more Smart Citation

Why Performance Models Matter for Grid Computing

Kennedy

IFIP the International Federation for Information Processing

View full text Add to dashboard Cite

Prion-Protein-Specific Aptamer Reduces PrPSc Formation

Proske¹,

Gilch²,

Wopfner³

et al. 2002

ChemBioChem

140

View full text Add to dashboard Cite

The critical initial event in the pathophysiology of transmissible spongiform encephalopathies (TSEs) appears to be the conversion of the cellular prion protein (PrP C ) into the abnormal isoform PrP Sc . This isoform forms high-molecular-weight protease K (PK) resistant aggregates that accumulate in the central nervous system of affected individuals. We have selected nuclease-resistant 2'-amino-2'-deoxypyrimidine-modified RNA aptamers which recognize a peptide comprising amino acid residues 90 ± 129 of the human prion protein with high specificity. This domain of prion proteins is thought to be functionally important for the conversion of PrP C into its pathogenic isoform PrP Sc and is highly homologous among prion proteins of various species including mouse, hamster, and man. Consequently, aptamer DP7 binds to the full-length human, mouse, and hamster prion protein. At low concentrations in the growth medium of persistently prion-infected neuroblastoma cells, aptamer DP7 significantly reduced the relative proportion of de novo synthesized PK-resistant PrP Sc within only 16 h. These findings may open the door towards a rational development of a new class of drugs for the therapy or prophylaxis of prion diseases.

show abstract

Scheduling strategies for mapping application workflows onto the grid

Cited by 132 publications

References 20 publications

Combined Fault Tolerance and Scheduling Techniques for Workflow Applications on Computational Grids

Combined Fault Tolerance and Scheduling Techniques for Workflow Applications on Computational Grids

Why Performance Models Matter for Grid Computing

Prion-Protein-Specific Aptamer Reduces PrPSc Formation

Contact Info

Product

Resources

About