End-to-End Delay Minimization for Scientific Workflows in Clouds under Budget Constraint

Wu, Chase Q.; Lin, Xiangyu; Yu, D.; Xu, Wei; Li, Li

doi:10.1109/tcc.2014.2358220

Cited by 89 publications

(37 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then, to further assess the efficiency of the previous algorithms, we made some comparisons with two competitors: Budget Distribution with Trickling (BDT [3]) and Critical Greedy (CG [25]). Both BDT and CG schedule deterministic workflows, and CG does not take into account communication costs.…”

Section: A Experimental Methodologymentioning

confidence: 99%

“…Both BDT and CG schedule deterministic workflows, and CG does not take into account communication costs. In [25], CG also comes with a refined version CG+. We extended BDT and CG/CG+ to fit our model, so as to enforce fair comparisons.…”

Section: A Experimental Methodologymentioning

confidence: 99%

“…To the best of our knowledge, the closest papers to this work are [3], [25], which both propose workflow scheduling algorithms (BDT in [3], CG/CG+ in [25]) under budget and deadline constraints, but with a simplified platform model. We extended BDT and CG/CG+ to enable a fair comparison with our algorithms, and present the corresponding results in Section V-D.…”

Section: Related Workmentioning

confidence: 99%

“…Then in Section V-C, we proceed to the refined variants HEFTBUDG+ and HEFTBUDG+INV, and quantify their improvement in terms of makespan, as well as their additional cost in terms of CPU time. Finally, Section V-D is devoted to comparing our algorithms with (extensions of) two formerly published competitors, namely BDT [3] and CG/CG+ [25]. for each T of LISTT do 7: for each host of (Used VM \sched[T ]) ∪ New VM do 8: sched ← schedule(T, host) 9: c tot , t calc,wf ← simulate(wf, P, LISTT, sched) 10: if ((t calc,wf < minT imeCalc) 11: and (c tot < B)) then 12: selectedHost ← host 13: minT imeCalc ← t calc,wf …”

Section: Simulationsmentioning

confidence: 99%

“…We describe budget-aware scheduling algorithms in Section IV: Section IV-A presents the extensions to MIN-MIN and HEFT, while Section IV-B provides the refined versions. Section V is devoted to assessing their performance through extensive simulations, including comparisons of two previous budget-aware algorithms, namely BDT [3] and CG/CG+ [25]. Finally, we provide concluding remarks and directions for future work in Section VI.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Budget-Aware Scheduling Algorithms for Scientific Workflows with Stochastic Task Weights on Heterogeneous IaaS Cloud Platforms

Caniou

Caron

Chang

et al. 2018

2018 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)

View full text Add to dashboard Cite

Abstract-This paper introduces several budget-aware algorithms to deploy scientific workflows on IaaS Cloud platforms, where users can request Virtual Machines (VMs) of different types, each with specific cost and speed parameters. We use a realistic application/platform model with stochastic task weights, and VMs communicating through a datacenter. We extend two wellknown algorithms, MIN-MIN and HEFT, and make scheduling decisions based upon machine availability and available budget. During the mapping process, the budget-aware algorithms make conservative assumptions to avoid exceeding the initial budget; we further improve our results with refined versions that aim at re-scheduling some tasks onto faster VMs, thereby spending any budget fraction leftover by the first allocation. These refined variants are much more time-consuming than the former algorithms, so there is a trade-off to find in terms of scalability. We report an extensive set of simulations with workflows from the Pegasus benchmark suite. Most of the time our budget-aware algorithms succeed in achieving efficient makespans while enforcing the given budget, despite (i) the uncertainty in task weights and (ii) the heterogeneity of VMs in both cost and speed values.

show abstract

Section: A Experimental Methodologymentioning

confidence: 99%

Section: A Experimental Methodologymentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Simulationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Budget-Aware Scheduling Algorithms for Scientific Workflows with Stochastic Task Weights on Heterogeneous IaaS Cloud Platforms

Caniou

Caron

Chang

et al. 2018

2018 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)

View full text Add to dashboard Cite

show abstract

Budget‐aware scheduling algorithms for scientific workflows with stochastic task weights on infrastructure as a service Cloud platforms

Caniou

Caron

Chang

et al. 2021

Concurrency and Computation

View full text Add to dashboard Cite

Summary This paper introduces several budget‐aware algorithms to deploy scientific workflows on Infrastructure as a Service Cloud platforms, where users can request Virtual Machines (VMs) of different types, each with specific cost and computing resources. We use a realistic application/platform model with stochastic task weights, and VMs communicating through a Cloud storage. We extend two well‐known algorithms, MinMin and HEFT, and make scheduling decisions based upon machine availability and remaining budget. During the mapping process, the budget‐aware algorithms make conservative assumptions to avoid exceeding the initial budget; we further improve the results with refined versions that aim at rescheduling some tasks onto faster VMs, thereby spending any budget fraction leftover by the first allocation. These refined variants are much more time‐consuming than the former algorithms, so there is a trade‐off to find in terms of scalability. We report an extensive set of simulations with workflows from the Pegasus benchmark suite. Most of the time, our budget‐aware algorithms succeed in achieving efficient makespans while enforcing the given budget, and this despite the uncertainty in task weights.

show abstract

Deep‐Q learning‐based heterogeneous earliest finish time scheduling algorithm for scientific workflows in cloud

et al. 2020

View full text Add to dashboard Cite

The complex and large-scale scientific workflow applications are effectively executes on the cloud. The performance of cloud computing highly depends on the task scheduling. Optimal workflow scheduling is still a challenge that needs to be addressed due to the conflicting objectives and increasing demand for quality of service. Task scheduling is an NP-hard problem due to its complexity.The newly introduced methods for resolving the problem of task scheduling are facing challenges to take the benefits of all aspects of cloud computing. In this article, we study the joint optimization of cost and makespan of scheduling workflows in infrastructure as a service clouds and propose a new workflow scheduling scheme using deep learning. In this scheme, a deep-Q learning-based heterogeneous earliest-finish-time (DQ-HEFT) algorithm is developed, which closely integrates the deep learning mechanism with the task scheduling heuristic HEFT. The workflowsim simulator is used for the experiment of the real-world and synthetic workflows. The experiment results demonstrate the efficiency of our proposed approach compared with existing algorithms. This technique can achieve significantly better makespan and speed metrics with a remarkably higher volume of data and can run faster compared with the existing workflow scheduling algorithms in cloud computing environment.

show abstract

End-to-End Delay Minimization for Scientific Workflows in Clouds under Budget Constraint

Cited by 89 publications

References 26 publications

Budget-Aware Scheduling Algorithms for Scientific Workflows with Stochastic Task Weights on Heterogeneous IaaS Cloud Platforms

Budget-Aware Scheduling Algorithms for Scientific Workflows with Stochastic Task Weights on Heterogeneous IaaS Cloud Platforms

Budget‐aware scheduling algorithms for scientific workflows with stochastic task weights on infrastructure as a service Cloud platforms

Deep‐Q learning‐based heterogeneous earliest finish time scheduling algorithm for scientific workflows in cloud

Contact Info

Product

Resources

About