An Economic-based Resource Management and Scheduling for Grid Computing Applications

Resource management and scheduling plays a crucial role in achieving high utilization of resources in grid computing environments. Due to heterogeneity of resources, scheduling an application is significantly complicated and challenging task in grid system. Most of the researches in this area are mainly focused on to improve the performance of the grid system. There were some allocation model has been proposed based on divisible load theory with different type of workloads and a single originating processor. In this paper we introduce a new resource allocation model with multiple load originating processors as an economic model. Solutions for an optimal allocation of fraction of loads to nodes obtained to minimize the cost of the grid users via linear programming approach. It is found that the resource allocation model can efficiently and effectively allocate workloads to proper resources. Experimental results showed that the proposed model obtained the better solution in terms of cost and time.

💡 Research Summary

The paper addresses the problem of resource management and scheduling in grid computing environments, focusing on the economic aspect of allocating workloads to heterogeneous resources. Traditional grid scheduling research has largely emphasized performance metrics such as makespan or resource utilization, often ignoring the monetary cost incurred by users. Moreover, most existing divisible‑load models assume a single source of work, which does not reflect realistic scenarios where multiple users submit jobs concurrently.

To overcome these limitations, the authors propose a novel resource allocation framework that supports multiple load‑originating processors (sources) and incorporates a cost model for each processing element. The grid is abstracted as a star‑topology: a central scheduler (broker) receives workloads L₁ … L_m from m users and distributes portions α_ij of each workload to n processing nodes P₁ … P_n. Each node j is characterized by three parameters: a communication rate z_j (time per unit of data transferred), a processing rate t_j (time per unit of computation), and a unit cost c_j (monetary cost per unit of work processed).

The objective is to minimize the total monetary expense incurred by all users while respecting deadlines and budget constraints. This is formulated as a linear programming (LP) problem (or mixed‑integer LP when binary selection variables x_ij are introduced). The key constraints are: (1) complete allocation of each source’s workload (∑_j α_ij = L_i); (2) execution of each assigned portion within the processor’s available time window