Load balancing scheduling mechanism for OpenStack and Docker integration

Abstract With the development of cloud-edge collaborative computing, cloud computing has become crucial in data analysis and processing. OpenStack Docker are important components integration two always attracted widespread attention industry. The scheduling mechanism adopted by existing fusion solution uses a uniform resource weight for all containers, nodes resources on platform is unbalanced under differentiated requirements containers. Therefore, considering different network communication qualities, load-balancing based proposed to meet needs various (CPU, memory, disk, bandwidth) This precise limitation strategy container centralized as basis, it generates exclusive weights containers through filtering stage, weighing stage adaptation, non-uniform memory access (NUMA) lean stage. experimental results show that, compared with Nova-docker Yun, reduces load imbalance within node 57.35% 59.00% average, average between reduced 53.53% 50.90%. can also achieve better balancing without regard bandwidth.