Resource Management in Real Time Distributed System with Security Constraints: A Review
International Journal of Distributed Systems and Technologies
| Hi-index | 0.01 |
Distributed computing environment consists of a collection of autonomous workstations connected by the LAN. Due to random arrival of tasks and their random CPU service time requirements, there is good possibility that several workstations are heavily loaded, while others are idle or lightly loaded. Many load sharing/balancing algorithms mainly target sharing CPU resources, and have been intensively evaluated in a homogeneous environment. The main purpose of our algorithm is to improve the overall response time of the distributed system with minimum network traffic over heterogeneous workstations. This algorithm tries to keep the load difference between any two nodes within specified thresholds, and shares CPU, Memory and IO (thus named IOCM\_LB) resources with the best effort. Theoretical analysis and experimental evaluation demonstrate that compared with existing approaches, proposed algorithm can reduce the mean slow down and network traffic.