Gmach have
studied the problem of the energy-efficient dynamic consolidation of VMs in enterprise environments. The authors
cyprus company have proposed a combination of a trace-based workload placement controller and a reactive migra- tion controller. The trace-based workload placement controller collects data on stationary bike stand resource usage by VMs instantiated in the data center and uses this historical information to optimize the allocation, while meeting the specified QoS require- ments. This controller performs multiobjective optimization by trying to find a new placement of VMs that will minimize the number of servers needed to serve the
workload, while limiting the number of VM migrations required to pet supplies achieve the new
placement. The bound on the number of migrations is supposed to be set by the system administrator depending on the acceptable VM migration overhead. The controller places VMs according to their peak resource usage over the period since the previous reallocation, which is set to 4 hours in the experimental study.
The reactive migration controller continuously
funny t shirts monitors the resource utilization of physical nodes and detects when the servers are overloaded or underloaded. In contrast to the trace-based workload placement controller, it acts based on the real- real estate time data on the resource usage and adapts the allocation in a small scale (every minute). The objective of this controller is to rapidly respond to fluctuations in the workload. The controller is parameterized by two utilization thresholds that deter- mine overload and underload conditions. An overloading occurs when the utilization of CPU or memory of a server exceeds a given threshold. An underloading occurs when the replica watches CPU or memory loveseat usage averaged over all the physical nodes falls below a specified threshold. The threshold values are statically set according to the workload analysis and QoS requirements.
The authors have proposed several policies Atkins Diet Food List based on different combinations of the described optimization controllers with different utilization thresholds. The simula- tion-driven evaluation using 3 months of real-world workload traces for 138 SAP applications has shown that the best results can be achieved by applying both optimi- zation controllers simultaneously. The best policy invokes the workload placement controller every 4 hours, and when the servers are detected to be lightly utilized. The migration controller is executed in parallel to tackle the overloading and underloading of servers when they occur. This policy provides minimal CPU violation penalties and requires 10–20% more CPU capacity than the ideal case.