Avoid Deadlock Resource Allocation (ADRA) Model V VM-out-of-N PM

  • Ha Huy Cuong Nguyen Department of Information Technology, Quang Nam University, Vietnam
  • Van Thang Doan Industrial University of Ho Chi Minh, Ho Chi Minh City, Vietnam
Keywords: Cloud computing, IaaS, Resource allocation, Heterogeneous, Distributed platforms, Avoid deadlock

Abstract

This paper presents an avoid deadlock resource allocation (ADRA) for model V VM-out-of-N PM since cloud computing is a new computing paradigm composed of grid computing, distributed computing and utility concepts. Cloud computing presents a different resource allocation paradigm than either grids or distributed systems. Cloud service providers dynamically scale virtualized computing resources as a service over the internet. Due to variable number of users and limited resources, cloud is prone to deadlock at very large scale. Resource allocation and the associated deadlock avoidance is problem originated in the design and the implementation of the distributed computing, grid computing. In this paper, a new concept of free space cloud is proposed to avoid deadlock by collecting available free resource from all allocated users. New algorithms are developed for allocating multiple resources to competing services running in virtual machines on a heterogeneous distributed platform.  An experiment is tested in CloudSim. The performance of resource pool manager is evaluated by using CloudSim and resource utilization and indicating good results.

Published
2019-01-01
How to Cite
[1]
H. H. C. Nguyen and V. T. Doan, “Avoid Deadlock Resource Allocation (ADRA) Model V VM-out-of-N PM”, Int. J. Innov. Technol. Interdisc. Sci., vol. 2, no. 1, pp. 98-107, Jan. 2019.