Effective Management of Data Centers Resources for Load Balancing in Cloud Computing

Effective Management of Data Centers Resources for Load Balancing in Cloud Computing

Pradeep Kumar Tiwari (Manipal University Jaipur, Rajasthan, India) and Sandeep Joshi (Manipal University Jaipur, Rajasthan, India)
Copyright: © 2018 |Pages: 17
DOI: 10.4018/IJIRR.2018040103
OnDemand PDF Download:
No Current Special Offers


An effective load balancing mechanism maximizes the throughput minimizes the response time with fault tolerance. Load imbalance problems occasionally occur during the over-demanding of resources from VMs. Load balance mechanisms manage the load by allocating jobs and reallocation of VMs. This article proposes the Dynamic Weighted Live Migration (DWLM) mechanism. DWLM mechanism based on LP-formulation-based heuristic approaches to dynamically manage load balancing. DWLM approaches use transfer, selection, and location polices. These policies work based on an information policy. The authors map the result in migration time, throughput, response time and fault tolerance. The proposed DWLM mechanism gives the best results from Equally Spared Current Execution (ESCEL) and Push-Pull mechanisms. A comparison table and associated charts show the efficiency of the proposed DWLM mechanism.
Article Preview


An effective Load balancing mechanism enhances the fair workload distribution among the VMs. Load balancing use the mechanism of hyper threading to use a single processor like multiple processors. Load balancing intends to minimize the resource and maximize the resource utilization. The core concept of effective load balancing technique maximizes the throughput and minimizes the response time with fault tolerance (Kaur & Luthra, 2012).

Cloud computing is the ability of using various computing resources through the internet including applications and storage services. The shared pool of the resources managed by virtualization component of Cloud computing. According to the National Institute of Standard Technology (NIST), Cloud Computing is defined as a model for providing convenient and on demand access to the shared pool of resources including Networks, Storage, and CPU etc. The load balancing mechanism is a key mechanism to manage the User Bases resources request from Data Centers (García-Galán, Trinidad, Rana, & Ruiz-Cortés, 2016).

Load balancing provide effective management of resources by resource allocation policy using the task scheduling in distributed environment. The load management mechanism should ensure the

  • Resource availability on time to reduce Service Level Agreement (SLA) violation

  • •Effective resource utilization during the high or low load.

  • •Cost effective by using effective management of resources.

  • •Increasing the Quality of Service (QoS) with robust fault tolerance mechanism

The resource allocation mechanism is based on fair distribution among the all available VMs. High loaded VMs jobs migrated to another VMs. Proper mapping of resources minimizes the resources and maximize the utilization of available resources. The load management, mapping feature is responsible for the allocation of cores and the division of processor by hyper-threading. And ensure the availability of resources to UB by VM resource utilization. Allocation policy is responsible for mapping the resources and management of overloaded condition (Hu, Zhao, Xu, Ding, & Chu, 2012).

The load balancing metrics are Throughput, Fault- Tolerance, Scalability, and Migration Time. These mapping parameters describe the efficiency and effectiveness of the DWLM load balancing mechanism.

Throughput outcome defines the number of tasks are executed in specific amount of time. Throughput indicates the capability of the system.

Fault tolerance is an ability of recovery from failure. Fault tolerance map with Availability and Reliability component.

Scalability is the mapping ability of the system with finite number of VMs.

Migration Time is the process time which is taken by load manager to migrate the job from one VM to another VM

Proposed DWLM approach results map on Throughput, Migration time, Scalability with Fault tolerance feature. DWLM approach symmetrically manages the Sender and Receiver mechanism. Sender initiative mechanism send the message the high load and receiver initiative send the message of low load to take the load form high loaded VM. DWLM also uses the Selection, Location, Transfer and Information policies.

Complete Article List

Search this Journal:
Volume 12: 4 Issues (2022): 3 Released, 1 Forthcoming
Volume 11: 4 Issues (2021)
Volume 10: 4 Issues (2020)
Volume 9: 4 Issues (2019)
Volume 8: 4 Issues (2018)
Volume 7: 4 Issues (2017)
Volume 6: 4 Issues (2016)
Volume 5: 4 Issues (2015)
Volume 4: 4 Issues (2014)
Volume 3: 4 Issues (2013)
Volume 2: 4 Issues (2012)
Volume 1: 4 Issues (2011)
View Complete Journal Contents Listing