Energy-efficient virtual machine placement in data centers with heterogeneous requirements

Abstract

The energy spent on powering the global infor mation technology (IT) infrastructure has grown tremendously in the past decade, despite the dramatic improvement achieved in building energy efficient IT devices. In particular, reducing the high energy consumption of data centers has become a very important area due to the sudden popularity and demand for cloud computing services. In this paper, we propose to optimize the energy consumption in data centers by placing customers' virtual machines intelligently in the data center. We formulate this problem as a mathematical optimization problem and prove it's NP-hardness. We further explore two greedy approximation algorithms: the minimum power virtual machine (VM) placement algorithm (MinPow), that attempts to decrease the power consumption by avoiding powering up unnecessarily additional servers; and, the minimum communication virtual machine placement algorithm (Min Com), that aims at packing one user's VMs as much as possible on the same server or under the same rack to decrease the networking energy cost.We further, examine the performance of these two algorithms in both small scale and large scale clusters, and compare them to other alternatives. Our results demonstrate that MinPow and MinCom yield placements that are within 5% to 10% energy consumption of the optimal solution.

Energy Efficient Virtual Cluster Embedding in Public Data Centers

“…For example Buyya et al [4] construct a framework to monitor the physical servers status and shut down servers when they are idle. Beloglazov et al [3] and Dai et al [5] concentrate on the VM allocation method. The former one uses exhaustive search method to achieve the best allocation of VMs.…”

“…Compared to VM scheduling in a multi-tenant environment [5], virtual cluster embedding is more challenging due to the additional requirement of incarnating the virtual topology that interconnects the VMs. Let K be the set of tenants.…”

“…Since the energy consumed by a server dynamically changes with its CPU load, we set the energy consumed by a server based on the number of CPUs, e.g., a server with 32 cores is configured with 144 × 4 Watts as full load consumption. The top-of-rack switches are configured with 48 ports and 1 linecard, other switches are configured with 5 if Sub RP is infeasible then 5: Reverse the allocation in V Mt v ; return False 6:…”

“…Researchers have proposed a wide spectrum of techniques to achieve energy efficiency, such as energy aware virtual machine (VM) scheduling [3]- [5] and energy aware virtual network embedding [6].…”

Energy-Efficient Virtual Machines Scheduling in Multi-Tenant Data Centers

“…In our previous work [22] we have considered a similar yet much simpler problem where the tenants require CPU cores, memory and disk space only. As such resources can be completely isolated by vitalization and the formulation resulted into a linear programming problem.…”