Power Minimization in Fat-Tree SDN Datacenter Operation

The problem of energy consumption in data centers has attracted many researchers interest recently. This paper proposes an optimal energy consumption Software Defined Network (SDN) data center model using the dynamic activation of hosts and switches. We model switches and hosts as queues and formulate a Mixed Integer Linear Programming (MILP) model to minimize energy consumption while guaranteeing Quality of Service (QoS) of data center. Our purpose is minimizing static power, port power, and memory power of data centers. Since the problem is NP-hard, we adopt Simulated Annealing algorithm to obtain the solution. Through numerical experiment, we could observe that our model is able to save reasonable energy compared to the full operation data center model.

[1]  Mohammed Samaka,et al.  Application delivery in multi-cloud environments using software defined networking , 2014, Comput. Networks.

[2]  Li Li,et al.  Joint power optimization of data center network and servers with correlation analysis , 2014, IEEE INFOCOM 2014 - IEEE Conference on Computer Communications.

[3]  Yongli Zhao,et al.  Data center optical networks (DCON) with OpenFlow based Software Defined Networking (SDN) , 2013, 2013 8th International Conference on Communications and Networking in China (CHINACOM).

[4]  C. DeCusatis,et al.  Dynamic, software-defined service provider network infrastructure and cloud drivers for SDN adoption , 2013, 2013 IEEE International Conference on Communications Workshops (ICC).

[5]  Kishor S. Trivedi,et al.  Stochastic Model Driven Capacity Planning for an Infrastructure-as-a-Service Cloud , 2014, IEEE Transactions on Services Computing.

[6]  Amin Vahdat,et al.  A scalable, commodity data center network architecture , 2008, SIGCOMM '08.

[7]  Changcheng Huang,et al.  Provisioning high-availability datacenter networks for full bandwidth communication , 2014, Comput. Networks.

[8]  Djamal Zeghlache,et al.  SDN for Inter Cloud Networking , 2013, 2013 IEEE SDN for Future Networks and Services (SDN4FNS).

[9]  Tao Zhang,et al.  Half-DRAM: A high-bandwidth and low-power DRAM architecture from the rethinking of fine-grained activation , 2014, 2014 ACM/IEEE 41st International Symposium on Computer Architecture (ISCA).

[10]  Dan Li,et al.  Software defined green data center network with exclusive routing , 2014, IEEE INFOCOM 2014 - IEEE Conference on Computer Communications.

[11]  Yi-Bing Lin,et al.  Waiting time distribution for the M/M/m queue , 2003 .