Self-Aggregation Techniques for Load Balancing in Distributed Systems

One of the today issues in software engineering is to find new effective ways to deal intelligently with the increasing complexity of distributed computing systems. In this context a crucial role is played by the balancing of the work load among all nodes in a system composed of interconnected nodes that enter and exit the system without following any rule. To address this issue, we are experimenting with the usage of autonomic self-aggregation techniques that rewire the system in groups of homogeneous nodes that are then able to balance the load among each others using classical techniques. We present our approach together with some simulation experiments that show how the application of self- aggregation algorithms makes it possible to balance the load also in these extreme situations. Besides, our experiments show that the introduction of self-aggregation does not introduce a significant overhead in terms of execution time, even if it requires the exchange of a higher number of messages between nodes.