A Knowledge Guided Multi-Population Evolutionary Algorithm for Dynamic Workflow Scheduling Problem

The workflow scheduling problem is a fundamental task in cloud computing. This paper addresses the challenge of workflow scheduling in dynamic and uncertain cloud environments, where computing resources may become inaccessible due to hardware or software failures. To tackle this challenge, we propose a novel algorithm called the Order Feature Guided Multi-Population (OFGMP) algorithm for dynamic workflow scheduling in cloud environments. The OFGMP algorithm utilizes a multi-population evolutionary framework, incorporating a knowledge-guided reproduction operator that leverages the order feature of solutions, as well as repair mechanisms to adapt to changing environmental conditions. Extensive experiments are conducted to validate the algorithm’s performance against existing dynamic scheduling approaches. The experimental results demonstrate the superiority of our proposed method over others on a number of test cases.