DocumentCode
3588636
Title
NICE: Network-aware VM Consolidation scheme for Energy Conservation in Data Centers
Author
Bo Cao ; Xiaofeng Gao ; Guihai Chen ; Yaohui Jin
Author_Institution
Dept. of Comput. Sci. & Eng., Shanghai Jiao Tong Univ., Shanghai, China
fYear
2014
Firstpage
166
Lastpage
173
Abstract
Energy conservation and network performance have become two of the most important issues in data center as the scale of cloud services continues growing. Recent researches usually consider these two issues separately. Energy conservation mainly deals with hosts, which reduces total energy consumption by consolidating virtual machine(VM)s to fewer hosts, and network performance mainly deals with network scalability and energy efficiency, which improves data center network(DCN) scalability by applying new network topologies or routing schemes and improves DCN energy efficiency by consolidating trafile. In this paper, we jointly consider these two issues and define Combined VM Consolidation (CVC) problem. We prove that CVC is NP-complete and is inapproximable by a factor of 3/2 ε unless P = NP. Next, we propose NICE: Network-aware VM Consolidation scheme for Energy Conservation in Data CEnter to solve CVC. Instead of taking the unrealistic hypothesis that migration cost is negligible, a common assumption in most literatures, we precisely analyze VM migration cost according to real-trace experiments in a 6-server testbed via VMware. Massive simulations validate the efficiency of NICE, In all, to the best of our knowledge, we arc the first work to combine VM consolidation with network optimization and migration cost.
Keywords
cloud computing; computational complexity; computer centres; optimisation; power aware computing; virtual machines; CVC; DCN; NICE; NP-complete problem; cloud service; combined VM consolidation; data center network; migration cost; network optimization; network-aware VM consolidation scheme for energy conservation in data centers; virtual machine; Nonvolatile memory; Silicon; VM consolidation; VM migration cost; cloud computing; data center; energy saving;
fLanguage
English
Publisher
ieee
Conference_Titel
Parallel and Distributed Systems (ICPADS), 2014 20th IEEE International Conference on
Type
conf
DOI
10.1109/PADSW.2014.7097805
Filename
7097805
Link To Document