Multi-Cloud Containerized Service Scheduling Optimizing Computation and Communication

Multi-cloud service scheduling is pivotal in optimizing Quality of Service (QoS). Among the key performance metrics for services, response time takes center stage, encompassing both computation and communication aspects. While previous research has primarily concentrated on enhancing communication efficiency, the computation component remains equally critical. In response to this, we propose a novel algorithm Multi-Cloud Service Scheduling Genetic Algorithm (MCSSGA) to schedule services in multi-cloud environments, taking into account service priorities. MCSSGA strives to minimize response times by optimizing CPU allocation and network latencies between services, addressing both computation and communication aspects. Experimental evaluations conducted within a real-world multi-cloud environment demonstrate that MCSSGA significantly outperforms the existing optimization policy that only focuses on communication. Specifically, MCSSGA accelerates the responses of 73.7% of services while maintaining robust service acceptance rates and ensuring reasonable scheduling times.