Resource Scheduling In Grid Computing

923 Words4 Pages
A Prioritized User Deadline Based and Cost Effective Resource Scheduling in

Grid Computing

Abstract: A Grid represents a decentralized interconnected parallel system with multiple heterogeneous

resources. Grid computing is a collection of computing resources that are collected from multiple

administrative domains to reach a common goal. By focusing on grid resource sharing and coordination,

managing capabilities, and attaining high efficiency, grid computing has become an important

component of the computer industry. Resource scheduling in computational grids has an important role

in improving the efficiency. The grid environment is very dynamic, with the number of resources, their

availability, CPU loads, and the amount of unused
…show more content…
A grid can be a collection of machines, sometimes referred to as nodes, resources,

members, donors, clients and hosts. They all contribute a combination of resources to the grid as a whole.

Some resources may be used by all users of the grid, while others may have some restrictions.

By focusing on grid resource sharing and coordination, managing capabilities, and attaining high

efficiency, grid computing has become an important component of the computer industry. However, it is

still in the developmental stage, and several issues and challenges remain to be addressed.

Of these issues and challenges, resource scheduling in computational grids has an important role

in improving the efficiency. The grid environment is very dynamic, with the number of resources, their

availability, CPU loads, and the amount of unused memory constantly changing. In addition, different

tasks have different characteristics that require different schedules. For instance, some tasks require high

processing speeds and may require a great deal of coordination between their processes. Finally, one of

the most important distinctive requirements of grid scheduling compared with other scheduling (such
…show more content…
It includes searching multiple administrative domains to

use a single machine or scheduling a single job for multiple resources use at a single site or multiple sites.

Grid scheduling is a software framework which collects resource state information and

requirements for execution of jobs, selects appropriate resources, predicts the potential performance for

each candidate schedule and determines the best schedule for the tasks to be executed on a resource,

subject to some performance goals [4].

Schedulers are also responsible for management of jobs such as allocating resources for jobs,

partitioning of jobs to perform parallel execution, data management, event correlation and service level

management capabilities. These schedulers can be arranged in a hierarchical structure with meta-

schedulers and local schedulers. The functions of the scheduler include: (1) collecting information of jobs

submitted to the grid system (2) collecting available resource information (3) computation of the mapping

of jobs to selected resources (4) allocation of jobs according to mapping and (5) monitoring the status of

job execution. The goal of scheduling is to achieve highest possible system throughput and to match
Open Document