Decentralized Load Balancing In Heterogeneous Systems Using Diffusion Approach

Abstract

The grid and cluster computing uses interconnected nodes to solve a problem in parallel in order to improve the response time of the system. Diffusive load balancing algorithms works well when the nodes in the system have the same processing capacity. But little attention was paid in diffusion load balancing algorithms in the literature for distributing the workload in the nodes with different processing capabilities when the load of the nodes is treated as integers. When the loads are distributed to the nodes without considering their processing capacities it would affect the response time of the system. Effective load balancing in heterogeneity can be achieved by considering the processing capacities of the nodes. This paper propose a diffusive load balancing algorithm which distributes a proportion of excessive workload of heavily loaded node to lightly loaded node by considering the nodes processing capabilities.

A small world based overlay network for improving dynamic load-balancing

“…The diffusion approach (Hu and Blake, 1999;Neelakantan, 2012) is a dynamic load-balancing technique where each node simultaneously sends the excessive workloads to its under loaded neighbors and receives workloads from its neighbors with higher workload (Boillat, 1990;Cybenko, 1989). In 1990, Boillat et al (Boillat, 1990) presented a new approach to solve the load balancing problem for parallel programs.…”

“…Previous studies have proposed numerous load-balancing algorithms targeting at static, small-scale, homogeneous and/or heterogeneous environments (Aakanksha and Bedi, 2007;Hu and Blake, 1999;Karger and Ruhl, 2004;Meyerhenke, 2009;Neelakantan, 2012;Yagoubi and Meddeber, 2010). The diffusion approach (Hu and Blake, 1999;Neelakantan, 2012) is a dynamic load-balancing technique where each node simultaneously sends the excessive workloads to its under loaded neighbors and receives workloads from its neighbors with higher workload (Boillat, 1990;Cybenko, 1989).…”

“…Our proposed algorithm uses the on-state information exchange (Neelakantan, 2012), a node broadcasts its information to only the set of neighbor-nodes whenever its status changes. In fact, the on-state information exchange strategy has the advantages of making more accurate decisions, however, the large overhead in communication makes it impractical for large systems (i.e.…”

“…We have implemented a discrete-event simulator using the SimJava (Howell and Mcnab, 1998) to compare the performance of our proposed approach with two of the most popular dynamic diffusion approaches, the nearest neighbor algorithm (Tada, 2011) and the original neighborhood algorithm (Neelakantan, 2012).…”

Advanced Diffusion Approach to Dynamic Load-Balancing for Cloud Storage

“…Presented here is a summary of work related to the approaches and techniques used in this paper [2,4,5,6,7]. In complex huge systems, it is not feasible for a node to collect the information of all other nodes in the system.…”