Cluster resource scheduling in cloud computing: literature review and research challenges

Khallouli, Wael; Huang, Jingwei

doi:10.1007/s11227-021-04138-z

Cited by 19 publications

(9 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Huang et al [72] examined the difficulties in resource demands, heterogeneity, fairness, and scalability in cluster scheduling for cloud computing. Promising adaptive solutions are provided by machine learning.…”

Section: Related Workmentioning

confidence: 99%

Contention-aware Greedy Heuristic Method and Learning based Method for Load Balancing through Scheduling for Containers in Cloud Computing Environments

Gogineni,

2024

Preprint

View full text Add to dashboard Cite

Containerization became indispensable in distributed environments for packaging software and dependencies in a lightweight executable container. In the era of big data and availability of cloud infrastructure, it is more so as distributed applications are more resource and data-intensive. Such High Performance Computing (HPC) applications are deployed in containerized services. However, data-intensive nature of those applications lead to poor performance unless scheduler considers it. In this paper, not only load balancing of containers but also performance of underlying applications is considered. Towards this end, a scheduling algorithm with unified optimization considering load balancing and also application performance is proposed. This algorithm, named Contention-aware Greedy Heuristic Scheduling and Load Balancing for Containers (CGHSLBC), helps in improving performance of containerization in distributed environments. Problems associated with containerization in terms of balancing load and also application performance are NP-hard. CGHSLBC has heuristics to deal with such issues. Empirical study has revealed that CGHSLBC better application performance besides balancing load of containerized services in cloud infrastructure. We also proposed a learning based methodology to schedule and balance load for containers. It is based on Deep Reinforcement Learning (DRL) where state change is continuously monitored while making well informed scheduling decisions. An algorithm named Reinforcement Learning based Dynamic Scheduling (RLbDS) is proposed and empirical study has revealed that it shows better performance over state of the art methods.

show abstract

Section: Related Workmentioning

confidence: 99%

Contention-aware Greedy Heuristic Method and Learning based Method for Load Balancing through Scheduling for Containers in Cloud Computing Environments

Gogineni,

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The updated formulas for the velocity and position of the particles are Eqs. ( 6) and (7), respectively.…”

Section: Enhanced Pso Algorithmmentioning

confidence: 99%

“…These redundant backups consume a portion of the energy. On the other hand, because the resources of the data center are not used reasonably and effectively, the resources are idle or wasted [7]. Many cloud computing service providers have begun to co-distribute batch jobs and online jobs on the same cluster in order to save economic costs and improve production efficiency.…”

Section: Introductionmentioning

confidence: 99%

A Multi-Swarm PSO Approach to Large-Scale Task Scheduling in a Sustainable Supply Chain Datacenter

Liu

Zeng

Liu

et al. 2022

Preprint

View full text Add to dashboard Cite

In recent years, supply chain management has become an incrementally vital part of industrial sectors, which affects the service quality and production efficiency of relevant enterprises. With the rapid development of cloud computing technology, a distributed datacenter plays a vital role in supply chain management of many infrastructures but suffers from high energy consumption and low service efficiency due to heavy allocation of massive calculation and analysis tasks. In this paper, an efficient multi-swarm particle swarm optimization approach is proposed based on load balancing. Both makespan and completion time variances of all resources have been minimized. An initialization scheme is also presented concerning the convergence rate and adaptive inertia weights. An open dataset from an Alibaba datacenter has been employed to resolve the uneven load events caused by inefficient supply chain arrangement between distributed tasks and resources. Two criteria, makespan and response time during task scheduling have been chosen for performance evaluation. According to the results, the proposed work can improve task scheduling efficiency and sustainability in distributed datacenters supporting diverse supply chain environments.

show abstract

“…A novel VM-assign load balancing technique that efficiently distributes requests among all accessible VMs has been presented in response to the load balancing issue [15]. Algorithms for scheduling cloud computing cluster resources have been reviewed in the literature [6]. Resource scheduling in IoT simulation frameworks is the study's main focus [2].…”

Section: Related Workmentioning

confidence: 99%

Implementing FCFS and SJF for finding the need of Reinforcement Learning in Cloud Environment

Lahande

Kaveri

2022

ITM Web Conf.

View full text Add to dashboard Cite

Cloud has grown significantly and has become a popular serviceoriented paradigm offering users a variety of services. The end-user submits requests to the cloud in the form of tasks with the expectation that they will be executed at the best possible lowest time, cost and without any errors. On the other hand, the cloud executes these tasks on the Virtual Machines (VM) by using resource scheduling algorithms. The cloud performance is directly dependent on how the resources are managed and allocated for executing the tasks. The main aim of this research paper is to compare the behaviour of cloud resource scheduling algorithms: First Come First Serve (FCFS) and Shortest Job First (SJF) by processing high-sized tasks. This research paper is broadly divided into four phases: the first phase includes an experiment conducted by processing approximately 80 thousand tasks from the Alibaba task event dataset using the resource scheduling algorithms: FCFS and SJF on the cloud VMs under different circumstances; the second phase includes the experimental results; the third phase includes a empirical analysis of the behaviour of resource scheduling algorithms; the last phase includes the proposed need of Reinforcement Learning (RL) to improve cloud resource scheduling and its overall performance.

show abstract

Cluster resource scheduling in cloud computing: literature review and research challenges

Cited by 19 publications

References 69 publications

Contention-aware Greedy Heuristic Method and Learning based Method for Load Balancing through Scheduling for Containers in Cloud Computing Environments

Contention-aware Greedy Heuristic Method and Learning based Method for Load Balancing through Scheduling for Containers in Cloud Computing Environments

A Multi-Swarm PSO Approach to Large-Scale Task Scheduling in a Sustainable Supply Chain Datacenter

Implementing FCFS and SJF for finding the need of Reinforcement Learning in Cloud Environment

Contact Info

Product

Resources

About