GreenSlice: Enabling renewable energy powered cellular base stations using asynchronous delivery

Globally, there is a tremendous growth in mobile data traffic and mobile network operators (MNO) are investing heavily to expand and operate their networks to serve them. The operational expenditure incurred by MNOs are increasing much more compared to revenue from mobile data traffic, thereby decreasing their profit margins. One of the major components of operational expenditure is the cost of operating cellular base stations using grid power or diesel generators in the absence of grid power. MNOs in off-grid sites constantly rely on diesel generators to run base stations that is very expensive. At the same time, they also cause adverse environmental impact due to CO2 emission. While the economics of renewable energy are already favorable in remote off-grid areas, it has been quite challenging for operators to handle the unpredictability of renewable energy sources while meeting the user QoE. In this work, we present GreenSlice, an `energy-source' aware asynchronous content delivery mechanism that adapts delivery of bulk delay-tolerant content in periods of availability of renewable energy. In particular, given a set of heavy demands (such as downloads of videos, software updates etc.), we present a scheduling algorithm that balances the operator's energy cost and delay experienced by users. We use a novel adaptation of Lyapunov optimization framework to provide bounds on the content delivery time and resulting operational cost. Simulation studies show our mechanism reduces the energy cost by more than 33% compared to traditional schemes of using renewables and non-renewables without asynchronous content delivery, at the same time keeping the increase in delay negligible.

[1]  Ulas C. Kozat,et al.  Dynamic resource allocation and power management in virtualized data centers , 2010, 2010 IEEE Network Operations and Management Symposium - NOMS 2010.

[2]  Vinay Kolar,et al.  Async: De-congestion and yield management in cellular data networks , 2013, ICNP.

[3]  Vincenzo Mancuso,et al.  Reducing costs and pollution in cellular networks , 2011, IEEE Communications Magazine.

[4]  Yuan Yao,et al.  Data centers power reduction: A two time scale approach for delay tolerant workloads , 2012, 2012 Proceedings IEEE INFOCOM.

[5]  Athanasios V. Vasilakos,et al.  A Survey of Green Mobile Networks: Opportunities and Challenges , 2012, Mob. Networks Appl..

[6]  Ness B. Shroff,et al.  Scheduling heterogeneous delay tolerant tasks in smart grid with renewable energy , 2012, 2012 IEEE 51st IEEE Conference on Decision and Control (CDC).

[7]  Vijay K. Bhargava,et al.  Green Cellular Networks: A Survey, Some Research Issues and Challenges , 2011, IEEE Communications Surveys & Tutorials.

[8]  Alexandros G. Dimakis,et al.  Efficient Algorithms for Renewable Energy Allocation to Delay Tolerant Consumers , 2010, 2010 First IEEE International Conference on Smart Grid Communications.

[9]  Michael J. Neely.  Stock Market Trading via Stochastic Network , 2010 .

[10]  Aaditeshwar Seth,et al.  Motivation and design of a content distribution architecture for rural areas , 2012, ACM DEV '12.

[11]  Ramesh Govindan,et al.  Energy-delay tradeoffs in smartphone applications , 2010, MobiSys '10.

[12]  Sangtae Ha,et al.  TUBE: time-dependent pricing for mobile data , 2012, SIGCOMM '12.

[13]  Prashant J. Shenoy,et al.  Predicting solar generation from weather forecasts using machine learning , 2011, 2011 IEEE International Conference on Smart Grid Communications (SmartGridComm).

[14]  Sumei Sun,et al.  Energy cooperation in cellular networks with renewable powered base stations , 2013, 2013 IEEE Wireless Communications and Networking Conference (WCNC).