Minimizing data center cooling and server power costs

Abstract

This paper focuses on power minimization in a data center accounting for both the information technology equipment and the air conditioning power usage. In particular we address the server consolidation (on/off state assignment) concurrently with the task assignment. We formulate the resulting optimization problem as an Integer Linear Programming problem and present a heuristic algorithm that solves it in polynomial time. Experimental results show an average of 13% power saving for different data center utilization rates compared to a baseline task assignment technique, which does not perform server consolidation.

Joint Computing and Electric Systems Optimization for Green Datacenters

“…A number of research works present methods for server consolidation based on per-VM workload characteristics, i.e., the peak, off-peak, and average utilization of workload [26,35]; which aim is to reduce heat dissipation of hot spot zones and improve overall power utilization in datacenters [9,20]. In [34] authors propose abstract models to balance computing power in a datacenter by minimizing peak inlet temperatures.…”

“…Server consolidation [26] is one of the widely used techniques to reduce the energy overheads, which minimizes the number of active servers by packing workloads or Virtual Machines (VMs) into the minimal number of active servers exploiting a virtualized environment. Large virtualized datacenters use renewable energy to reduce their dependence on costly and brown energy from the grid [33].…”

Power-Thermal Modeling and Control of Energy-Efficient Servers and Datacenters

“…φ i,j indicates the contribution of the outlet heat rate of the i-th server in the inlet heat rate of the j -th one. Assuming Q out i and Q in j are, respectively, the outlet and inlet heat rates for the i-th and j -th server, the inlet heat rate for j -th server can be calculated as follows [39]:…”

“…Innovative DCIM support systems for datacenter management are thus needed. PMSM (i.e., Power Monitor System and Management) [44][45][46][47][48][49][50][51][52][53][54][55][56], developed at EPFL in cooperation with Credit Suisse [45], is an example of such an innovation.…”

“…3, the power consumption of servers is not proportional to the utilization, i.e., the idle power consumes around 50 % of the peak power consumption. Due to the poor energyproportionality, the power consumed by servers in datacenters can be reduced as we minimizes the number of active servers by packing workloads into the minimal number of active servers [46]. The technique is called Server consolidation.…”