Modelling the power consumption and trade-offs of virtualised cloud radio access networks

In large-scale computing centres, the advancement of knowledge in regard to the predicted power consumption (PC) and concerns of host servers that run virtual machines (VMs) could improve the capacity planning and networks' energy efficiency. A parameterised power model is proposed to explore the individual components within the virtualisation-based cloud radio access network. The model evaluates the PC and trade-offs of a server undergoing virtualisation. After cooling and total PC for cloud radio access network architecture, with and without virtualisation have been compared using differentiated parameters, such as varying number of bare-metal base band units, VMs and system's resource blocks (RBs) share/bandwidth. The results show dramatic decrease in the total PC via virtualising the core network. In addition, the degraded performance of each virtualised server is demonstrated via modelling the execution time and overhead costs. These costs have been resulted from increasing the number of hosted VMs and utilised RBs by each VM.

[1]  Muhammad Ali Imran,et al.  How much energy is needed to run a wireless network? , 2011, IEEE Wireless Communications.

[2]  Kyu Ho Park,et al.  Credit-Based Runtime Placement of Virtual Machines on a Single NUMA System for QoS of Data Access Performance , 2015, IEEE Transactions on Computers.

[3]  Harald Haas,et al.  A Parameterized Base Station Power Model , 2013, IEEE Communications Letters.

[4]  Sampath Rangarajan,et al.  Radio access network virtualization for future mobile carrier networks , 2013, IEEE Communications Magazine.

[5]  Min Chen,et al.  Software-Defined Network Function Virtualization: A Survey , 2015, IEEE Access.

[6]  Vera Stavroulaki,et al.  5G on the Horizon: Key Challenges for the Radio-Access Network , 2013, IEEE Vehicular Technology Magazine.

[7]  John H. Hartman,et al.  Energy-efficient memory management in virtual machine environments , 2011, 2011 International Green Computing Conference and Workshops.

[8]  Qiang Huang,et al.  Power Consumption of Virtual Machine Live Migration in Clouds , 2011, 2011 Third International Conference on Communications and Mobile Computing.

[9]  Hai Jin,et al.  Poris: A Scheduler for Parallel Soft Real-Time Applications in Virtualized Environments , 2016, IEEE Transactions on Parallel and Distributed Systems.

[10]  Navid Nikaein,et al.  5G Architectural Design Patterns , 2016, 2016 IEEE International Conference on Communications Workshops (ICC).

[11]  Hamed S. Al-Raweshidy,et al.  Component and parameterised power model for cloud radio access network , 2016, IET Commun..

[12]  M. Zhang,et al.  Hierarchical virtual network mapping algorithm for large-scale network virtualisation , 2012, IET Commun..

[13]  Marius Marcu,et al.  Power consumption measurements of virtual machines , 2011, 2011 6th IEEE International Symposium on Applied Computational Intelligence and Informatics (SACI).

[14]  Yong Li,et al.  System architecture and key technologies for 5G heterogeneous cloud radio access networks , 2015, IEEE Netw..

[15]  Ricardo Lent,et al.  Evaluating the Performance and Power Consumption of Systems with Virtual Machines , 2011, 2011 IEEE Third International Conference on Cloud Computing Technology and Science.

[16]  Hamed S. Al-Raweshidy,et al.  Evaluating the energy efficiency of software defined-based cloud radio access networks , 2016, IET Commun..

[17]  Vikram Srinivasan,et al.  CloudIQ: a framework for processing base stations in a data center , 2012, Mobicom '12.

[18]  Ryan Shea,et al.  Power consumption of virtual machines with network transactions: Measurement and improvements , 2014, IEEE INFOCOM 2014 - IEEE Conference on Computer Communications.

[19]  Yonggang Wen,et al.  Data Center Energy Consumption Modeling: A Survey , 2016, IEEE Communications Surveys & Tutorials.