Optimize hardware power draw: Processors of physical machines can adapt their levels of voltage and frequency based on the levels of power consumed.26 This is known as dynamic voltage and frequency scaling (DVFS). DVFS reduces electricity costs for powering servers, as well as their thermal output, lowering the energy demand for cooling.
Automate server consolidation: Data center operators can enhance power efficiency through consolidation—minimizing the number of active servers. For virtualized environments, operators can run a workload orchestrator that monitors the utilization of the physical hosts, then puts to sleep underutilized hosts while redirecting inbound requests to the VMs running on the active hosts.27
Optimize rack space: Rack space is generally cheap compared to hardware, so data center operators might not pay much attention to it. But it can have an impact for GreenOps. According to Jeff Riggen, Stratascale’s Senior Technical Advisor—Hybrid Cloud: “The leasing of rack space is becoming more common. The more intensively you use rack space, the less energy inefficient things are.”28
Implement applied scheduling: The applied scheduling method assigns users’ requests for services to VMs within the boundaries of their service level agreement (SLA).29 To achieve the targeted reductions in power consumption, the applied scheduling method then selects VMs on the servers that are capable of consuming less power.30