More than Meets One Core: An Energy-Aware Cost Optimization in Dynamic Multi-Core Processor Server Consolidation for Cloud Data Center

The massive number of users has brought severe challenges in managing cloud data centers (CDCs) composed multi-core processor that host service providers. Guaranteeing the quality (QoS) multiple as well reducing operating costs CDCs are major problems need to be solved. To solve these problems, this paper establishes a cost model based on hosts CDCs, which comprehensively consider hosts’ energy costs, virtual machine (VM) migration and level agreement violation (SLAV) penalty costs. optimize goal, we design following solution. We employ DAE-based filter preprocess VM historical workload use an SRU-based method predict computing resource usage VMs future periods. Based predicted results, trigger migrations before move into overloaded state reduce occurrence SLAV. A multi-core-aware heuristic algorithm is proposed placement problem. Simulations driven by real dataset validate effectiveness our method. Compared with existing baseline methods, reduces total 20.9~34.4%.