Limits...
Job Scheduling with Efficient Resource Monitoring in Cloud Datacenter.

Loganathan S, Mukherjee S - ScientificWorldJournal (2015)

Bottom Line: Since the resources are limited in these private clouds maximizing the utilization of resources and giving the guaranteed service for the user are the ultimate goal.For that, efficient scheduling is needed.This research reports on an efficient data structure for resource management and resource scheduling technique in a private cloud environment and discusses a cloud model.

View Article: PubMed Central - PubMed

Affiliation: Department of Information Science and Technology, Anna University, Tamil Nadu 600025, India.

ABSTRACT
Cloud computing is an on-demand computing model, which uses virtualization technology to provide cloud resources to users in the form of virtual machines through internet. Being an adaptable technology, cloud computing is an excellent alternative for organizations for forming their own private cloud. Since the resources are limited in these private clouds maximizing the utilization of resources and giving the guaranteed service for the user are the ultimate goal. For that, efficient scheduling is needed. This research reports on an efficient data structure for resource management and resource scheduling technique in a private cloud environment and discusses a cloud model. The proposed scheduling algorithm considers the types of jobs and the resource availability in its scheduling decision. Finally, we conducted simulations using CloudSim and compared our algorithm with other existing methods, like V-MCT and priority scheduling algorithms.

No MeSH data available.


Representation of data structure as linked list of a host.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4587428&req=5

fig2: Representation of data structure as linked list of a host.

Mentions: This component monitors and gathers information of a host such as running job (VM), number of executing cores, free core availability, and assigned AR jobs. This component prepares a resource availability list (RAL) and a preemption List (PL) from the information it got from the hosts. Essentially the RAL contains list of tuples <Host ID, CurrentTime, Free core availability, Free memory availability, Earliest Core available, Earliest Available Time (EAT)> and PL contains <Host ID, Job ID, Number of cores assigned, Time interval, Flag status>. Earliest available time (EAT) can be calculated using (1) and (3) below. To prepare the list, the component calculates the EAT of a core and the number of available cores for a time interval and updates the list whenever a new job is assigned to a host or when a job completes releasing resources. We observe that CMS needs to refer to and use the information about the available resources from RAL and PL. Hence it is important that these lists are kept updated. We propose to use an appropriate data structure using which the two lists can be updated without delay. The proposed data structure aids in the search, retrieval, and updating mechanisms that take place. Figure 2 shows the modified partial data structure as linked list [16] is used in this research, which accommodates different job types. We identify the following basic operations to be performed by the data structure:


Job Scheduling with Efficient Resource Monitoring in Cloud Datacenter.

Loganathan S, Mukherjee S - ScientificWorldJournal (2015)

Representation of data structure as linked list of a host.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4587428&req=5

fig2: Representation of data structure as linked list of a host.
Mentions: This component monitors and gathers information of a host such as running job (VM), number of executing cores, free core availability, and assigned AR jobs. This component prepares a resource availability list (RAL) and a preemption List (PL) from the information it got from the hosts. Essentially the RAL contains list of tuples <Host ID, CurrentTime, Free core availability, Free memory availability, Earliest Core available, Earliest Available Time (EAT)> and PL contains <Host ID, Job ID, Number of cores assigned, Time interval, Flag status>. Earliest available time (EAT) can be calculated using (1) and (3) below. To prepare the list, the component calculates the EAT of a core and the number of available cores for a time interval and updates the list whenever a new job is assigned to a host or when a job completes releasing resources. We observe that CMS needs to refer to and use the information about the available resources from RAL and PL. Hence it is important that these lists are kept updated. We propose to use an appropriate data structure using which the two lists can be updated without delay. The proposed data structure aids in the search, retrieval, and updating mechanisms that take place. Figure 2 shows the modified partial data structure as linked list [16] is used in this research, which accommodates different job types. We identify the following basic operations to be performed by the data structure:

Bottom Line: Since the resources are limited in these private clouds maximizing the utilization of resources and giving the guaranteed service for the user are the ultimate goal.For that, efficient scheduling is needed.This research reports on an efficient data structure for resource management and resource scheduling technique in a private cloud environment and discusses a cloud model.

View Article: PubMed Central - PubMed

Affiliation: Department of Information Science and Technology, Anna University, Tamil Nadu 600025, India.

ABSTRACT
Cloud computing is an on-demand computing model, which uses virtualization technology to provide cloud resources to users in the form of virtual machines through internet. Being an adaptable technology, cloud computing is an excellent alternative for organizations for forming their own private cloud. Since the resources are limited in these private clouds maximizing the utilization of resources and giving the guaranteed service for the user are the ultimate goal. For that, efficient scheduling is needed. This research reports on an efficient data structure for resource management and resource scheduling technique in a private cloud environment and discusses a cloud model. The proposed scheduling algorithm considers the types of jobs and the resource availability in its scheduling decision. Finally, we conducted simulations using CloudSim and compared our algorithm with other existing methods, like V-MCT and priority scheduling algorithms.

No MeSH data available.