A cloud resource management framework for multiple online scientific workflows using cooperative reinforcement learning agents

Asghari, Ali; Sohrabi, Mohammad Karim

doi:10.1016/j.comnet.2020.107340

Cited by 33 publications

(11 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In 2020, Ali Asghari et al [7] developed a new architecture made up of many cooperating agents that took into account all aspects of TS and resource provisioning and managed the QoS offered to users. The integrated model that was suggested included all processes for TS and resource provisioning, and its many components help with managing user applications and making better use of cloud resources.…”

Section: Related Workmentioning

confidence: 99%

“…The scheduling solution was optimized in terms of each metric specified in the QoS model using a QoS-driven CESS method [4]. It has been suggested to use an OP-MLB system, which combines a number of algorithms that cooperate to provide effective resource management for cloud environments [6] [7]. With more effective storage and V/F scaling improvement, the EARU model significantly reduces LLC disappointments and thus more effectively utilizes asset.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An AHP based Task Scheduling and Optimal Resource Allocation in Cloud Computing

Karimunnisa¹,

Yellamma²

2023

IJACSA

View full text Add to dashboard Cite

Cloud systems by virtue characterize ultimate resource utilization with ever evolving user requirements facilitating adaptivity. With a scope of enhancing the QoS needs of user applications, numerous factors are considered for tunning among which Task scheduling promises to grab focus. The Task Scheduling mechanism ascertains improvement by distributing the subtasks to specific set of resources pertaining to prevailing Quality models. The work emphasizes the need for effective task scheduling and optimizing resource allocation by modelling a modified AHP (Analytical Hierarchy Process) driven approach. The proposed method guarantees the functionality in two phases pertaining to Task ranking and pipelined with Optimized scheduling algorithms resulting in maximization of resource utilization. The former phase of task ranking is aided by improved AHP with substantial usage of fuzzy clustering followed by an enhanced CUCMCA (Chimp Updated and Cauchy Mutated Coot Algorithm) algorithm for optimal resource allocation of cloud applications. The contributed model promises leveraged performance of 32% for memory usage, 33.5% for execution time, 29% for makespan and 18% for communication cost over pre-existing conventional models considered.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

An AHP based Task Scheduling and Optimal Resource Allocation in Cloud Computing

Karimunnisa¹,

Yellamma²

2023

IJACSA

View full text Add to dashboard Cite

show abstract

“…Accordingly, these methods have recently been used for multi-constraint process scheduling problems. Examples include the use of the sequential cooperative game multi-objective algorithm to improve cost and makespan for scheduling workflow in large scale presented by Duan et al, 23 the use of the multiple workflow scheduling algorithm using simple single-agent Q-learning to reduce the implementation of cloud tasks proposed by Cui et al, 24 the use of the random reinforcement learning-based task scheduling scheme with multiple objectives to increase resource utilization, improve load balancing, and reduce response time to tasks introduced by Peng et al, 25 the use of deep reinforcement learning in a hierarchical framework for energy management and cloud resource allocation method presented by Liu et al, 26 the use of the multi-swarm multi-objective optimization algorithm (MSMOOA) for the workflow scheduling in cloud computing for makespan, cost, and energy consumption introduced by Yao et al, 27 the use of the static task scheduling algorithm to reduce makespan with the title of QL-HEFT learning combination Q-learning with heterogeneous earliest finish time algorithm presented by Tong et al, 28 the use of multi-agent reinforcement learning based on deep-Q-network (DQN) for the workflow scheduling parallel to reduce cost and optimize workflow finish time provided by Wang et al, 29 the use of a framework based on online reinforcement learning to schedule data center (DC) task in warehouse-scale suggested by Cheng et al, 30 the use of the cloud resource management algorithm using free reinforcement learning models and fuzzy state estimation methods to reduce response time and using VM in 5G network presented by Jin et al, 31 the DQN algorithm for an online resource scheduling to optimize energy consumption and task makespan by Peng et al, 32 and the use of the cloud resource management structure for multiple online scientific workflows with several cooperative agents proposed by Asghari et al 33…”

Section: Agent-based Algorithmsmentioning

confidence: 99%

A cooperative multi‐agent offline learning algorithm to scheduling IoT workflows in the cloud computing environment

Gholami

Rezvan

2022

Concurrency and Computation

View full text Add to dashboard Cite

Regarding the problem of workflow scheduling in cloud environments, users want the workflow to be processed at a suitable time while cloud providers want to increase resource utilization. This article proposes a cooperative multi-agent offline learning algorithm called CMOL for minimizing makespan and energy consumption. This algorithm schedules a workflow that is represented by a directed acyclic graph (DAG) and assigns them to virtual machines (VMs). Multiple parallel agents interact and cooperate based on an algorithm in three steps of research, improvement, and selection to meet the imposed constraints of deadline and energy. Depending on the number of DAG levels, there is the same number of specialist agents who use strategies to create a Pareto feasible solution and simultaneously gain experience in the first two steps. The parallel agents exploit the extracted knowledge to improve the solution obtained by ensembling their experience in the selection step. To compare the efficiency of CMOL, two algorithms based on multi-agent systems and one algorithm based on single-agent are developed. The performance of the four algorithms is investigated on different real-world workflows and compared on various sizes. Computational results reveal the competitiveness of CMOL and its relative superiority compared with others.

show abstract

“…For small organizations involved in the federation, this process is especially important as it gives them the ability to expand their business with minimal resources. Resource management: work scheduling and resource provision among locals are a few of the main problems for CSPs [10], [11]. Monitoring of resources may help system administrators monitor cloud computing platform resources.…”

Section: Introductionmentioning

confidence: 99%

Monitoring and resource management taxonomy in interconnected cloud infrastructures: a survey

et al. 2022

View full text Add to dashboard Cite

Cloud users have recently expanded dramatically. The cloud service providers (CSPs) have also increased and have therefore made their infrastructure more complex. The complex infrastructure needs to be distributed appropriately to various users. Also, the advances in cloud computing have led to the development of interconnected cloud computing environments (ICCEs). For instance, ICCEs include the cloud hybrid, intercloud, multi-cloud, and federated clouds. However, the sharing of resources is not facilitated by specific proprietary technologies and access interfaces used by CSPs. Several CSPs provide similar services but have different access patterns. Data from various CSPs must be obtained and processed by cloud users. To ensure that all ICCE tenants (users and CSPs) benefit from the best CSPs, efficient resource management was suggested. Besides, it is pertinent that cloud resources be monitored regularly. Cloud monitoring is a service that works as a third-party entity between customers and CSPs. This paper discusses a complete cloud monitoring survey in ICCE, focusing on cloud monitoring and its significance. Several current open-source monitoring solutions are discussed. A taxonomy is presented and analyzed for cloud resource management.

show abstract

A cloud resource management framework for multiple online scientific workflows using cooperative reinforcement learning agents

Cited by 33 publications

References 68 publications

An AHP based Task Scheduling and Optimal Resource Allocation in Cloud Computing

An AHP based Task Scheduling and Optimal Resource Allocation in Cloud Computing

A cooperative multi‐agent offline learning algorithm to scheduling IoT workflows in the cloud computing environment

Monitoring and resource management taxonomy in interconnected cloud infrastructures: a survey

Contact Info

Product

Resources

About