Improved Load Balancing Algorithm for a Novel Distributed Computing Framework Based on MultiObjective Optimization Algorithm.

Ray is a novel distributed computing framework designed for artificial intelligence computations and has been widely applied in various fields. However, its default load balancing algorithm, which is based on the equal distribution approach, often leads to load imbalance issues. In this paper, we propose a load balancing algorithm based on multi-objective optimization to address the load imbalance problem in Ray. The algorithm utilizes an ideal similarity-based sorting solution to improve the shortcomings of Ray’s default load balancing algorithm. Experimental results demonstrate that the proposed algorithm effectively improves the load balancing performance.