Dynamic optimization of workload on compute nodes in private, public and hybrid clouds

Abstract

This system is designed for automatic workload distribution in a cluster by analysing workload of compute nodes and migrating VMs from overloaded nodes to underloaded ones. Besides workload stabilization, the system also provides an opportunity to reduce power consumption by unloading and suspending underloaded nodes. Workload stabilization in a cluster increases stability and reduces system response time.

Keywords

• балансировка
• кластер
• облачные вычисления