Fairness in dataflow scheduling in the cloud

Pietri, Ilia; Chronis, Yannis; Ioannidis, Yannis

doi:10.1016/j.is.2019.03.003

Cited by 8 publications

(4 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The global optima are passed to the next generation via a target immigration operator to maintain global optima during the algorithm repetitions. In the recent work, 29 a static scheduling approach was introduced, and it works on multiple DAGs which a single user submits. In this work, Pareto optimal schedules were defined in three‐dimensional space on heterogeneous clouds.…”

Section: Related Workmentioning

confidence: 99%

A new reliability‐based task scheduling algorithm in cloud computing

Motlagh

Movaghar

Rahmani

2021

Int J Communication

View full text Add to dashboard Cite

Summary In the last decade, the scale of heterogeneous computing (HC) systems such as heterogeneous cloud computing environments was growing like never before. So network failures are unavoidable in such systems, which affect system reliability. Since the task scheduling algorithm in HC is challenging, we investigate a new reliability‐aware task scheduling algorithm (RATSA) in this paper. RATSA is designed to schedule tasks on directed acyclic graphs (DAGs) by using the shuffled frog‐leaping algorithm (SFLA) and genetic algorithm (GA) as evolutionary algorithms. The population‐based SFLA‐GA is applied to optimize makespan in the RATSA as an NP‐complete problem. Moreover, the proposed algorithm exploits a new heuristic‐based earliest finish time technique for task mapping to virtual machines (VMs) section to decrease the failure rate. Experimental results on random DAGs indicate that RATSA improves some current algorithms in terms of reliability and has acceptable performance in makespan. The results reveal that the RATSA decreases the overall failure rate by 43% compared to some current task scheduling algorithms.

show abstract

Section: Related Workmentioning

confidence: 99%

A new reliability‐based task scheduling algorithm in cloud computing

Motlagh

Movaghar

Rahmani

2021

Int J Communication

View full text Add to dashboard Cite

show abstract

“…The goal was the efficient allocation of tasks to the different resources for meeting the real-time requirement of the cloud and edge computing environments. Pietri et al [40] proposed an algorithm that seeks to identify Pareto-optimal trade-offs between overall execution time, monetary cost and fairness, efficiently exploring the solution space. The authors approached the scheduling multiple dataflows on heterogeneous clouds.…”

Section: Related Workmentioning

confidence: 99%

“…In those situations, the scheduling by FIFO tends to concentrate on the workflow initial tasks, and it ends up degrading execution performance of integration processes. It becomes necessary a fair task scheduling, which provides optimal resources allocation to maximise performance [8,40]. Such allocation of optimal resources requires an active and dynamic task scheduling heuristic [51], which is capable of increasing the number of processed messages per time unit.…”

Section: Introductionmentioning

confidence: 99%

Queue-priority optimized algorithm: a novel task scheduling for runtime systems of application integration platforms

et al. 2021

View full text Add to dashboard Cite

The need for integration of applications and services in business processes from enterprises has increased with the advancement of cloud and mobile applications. Enterprises started dealing with high volumes of data from the cloud and from mobile applications, besides their own. This is the reason why integration tools must adapt themselves to handle with high volumes of data, and to exploit the scalability of cloud computational resources without increasing enterprise operations costs. Integration platforms are tools that integrate enterprises' applications through integration processes, which are nothing but workflows composed of a set of atomic tasks connected through communication channels. Many integration platforms schedule tasks to be executed by computational resources through the First-in-first-out heuristic. This article proposes a Queue-priority algorithm that uses a novel heuristic and tackles high volumes of data in the task scheduling of integration processes. This heuristic is optimized by the Particle Swarm Optimization computational method. The results of our experiments were confirmed by statistical tests, and validated the proposal as a feasible alternative to improve integration platforms in the execution of integration processes under a high volume of data.

show abstract

“…Building fair scheduling while increasing performance is a significant concern of enterprises, once they concurrently submit workflows for execution in different resources. Pietri et al [21] define fair scheduling as one that provides an adequate balancing from the workload to resources. In the domain of application integration, to obtain fair scheduling in contemporaneous environments, it is required to re-engineer integration platforms [22].…”

Section: Introductionmentioning

confidence: 99%

A Multi-queue Round Robin Algorithm for Task Scheduling in Enterprise Application Integration Platforms in the Cloud

Freire

Frantz

Basto‐Fernandes

et al. 2023

Preprint

View full text Add to dashboard Cite

Contemporaneous environments involve a large number of information which is received from many interconnected devices and integrated by many platforms. The Internet of Things and Cloud computing offer services to deal with such big data, so that enterprises can play in their business core by outsourcing the development, the configuration, and the maintenance of those services and their computational resources. Many enterprises use integration Platform as a Service to maintain their applications and data working in a synchronous way. However, current task scheduling algorithms of integration platforms are facing some hardships to tackle with high volumes of data. In this article, we propose a scheduling algorithm based in the round-robin heuristic through multiple task queue, which presents better performance than the traditional First-in-first-out heuristic used by current platforms. We experimented our algorithm in the execution of an integration process, and we validated the results with statistical techniques.

show abstract

Fairness in dataflow scheduling in the cloud

Cited by 8 publications

References 4 publications

A new reliability‐based task scheduling algorithm in cloud computing

A new reliability‐based task scheduling algorithm in cloud computing

Queue-priority optimized algorithm: a novel task scheduling for runtime systems of application integration platforms

A Multi-queue Round Robin Algorithm for Task Scheduling in Enterprise Application Integration Platforms in the Cloud

Contact Info

Product

Resources

About