Efficient Load balancing to serve Heterogeneous Requests in Clustered Systems using Kubernetes

Load balancing is an important part of a distributed computing environment which ensures that all devices or processors perform the same amount of work in an equal amount of time. Most load balancing algorithms assume similar processing power and workload for all the processors. However, now systems have become more complex and can have processors of different capabilities, workload, and configurations. In this paper, we propose an alternative algorithm for scheduling tasks. We configure the clusters dedicated to a particular type of task(real-time, dataintensive, etc.). Labels have been defined for each job to classify them into these categories. Then we modify the algorithm to introduce load balancing techniques using task migration.