Green Analysis for Content Centric Networking (CCN)

Overview


Green networking is a topic that is gaining importance at a rapid pace. Energy-efficiency and carbon-footprint are considered to be important parameters in designing a new networking architecture. Content-Centric Networking (CCN) [1] is a new networking architecture proposed for the future Internet. It has the potential to provide better results in terms of bandwidth usage, scalability and security as compared to the current IP-based architecture. In this project we try to evaluate the performance of CCN as compared to IP-based network in a video streaming application with regard to green networking.

Description


This project is divided into the following sections

  • Energy-assessment analysis.
  • Carbon-footprint analysis.

Energy-Assesment Analysis

In this analysis we present a intuitive model for content dissemination in CCN and conduct an energy consumption analysis of it compared to IP-based network in a video streaming scenario. We consider two types of energy consumption. First, we calculate the emergy (i.e., the energy required to manufacture) of the network devices [2]. Second, we calculate the energy required to operate the network devices for streaming the video. CCN network devices (routers) have a higher emergy and require more energy to operate compared to the IP-based network devices, however by exploiting their caching capability substantial energy benefits can be reaped. The emergy is a one time cost and does not change once the network is deployed. It is quite evident that in a CCN architecture, routers at higher levels have smaller link rates than routers at lower levels (due to caching of popular content). We implemented a practical CCN simulator ( motivated from the work of Carofiglio et al. in [3] ) to affirm this feature. We exploit this important feature by using routers which are capable of changing their operating rate according to the rate variation on a given link a phenomenon known as Rate Adaptation (RA). 

We consider three scenarios for our energy analysis. Ideal RA (with RA), practical RA, without RA.

Results

                                 

                                                               (a)                                                                                                                   (b)

Figure (a) shows that higher levels in a general tree topology (binary for this case), CCN with ideal RA is energy efficient. The figure show that the energy savings at the higher levels of the tree are more pronounce. Even if we do not use RA the power ratio (PIP /PCCN) is very close to 1, indicating that the energy cost associated with manufacturing and powering the additional cache memory is relatively small. The practical RA implementation based on provisioning according to demand ensures energy gain for CCN at higher levels in the network but it has poorer behavior at lower levels.

Figure (b) shows that it is possible to find a size of the cache for the CCN routers that maximizes the power ratio. For the specific settings in our simulations 128 GB appears to maximize the power efficiency.

 


Carbon footprint Analysis

We take the motivation from the work of Seetharam et al. [4] and estimate the carbon footprint values for our analysis. Carbon footprint of CCN and IP-based networks can be determined by calculating the amount of carbon dioxide emitted due to manufacturing and operation (usage) of network equipment used in both kind of networks.

Results

 

The carbon footprint due to manufacturing in case of CCN is higher as compared to IP. This is due to the presence of cache memory in CCN routers. CCN only out performs IP in terms of lesser carbon dioxide emission if we consider ideal RA. In case of practical and no RA IP produces less carbon dioxide.

 

Poster


A poster was presented on the topic at  Colloquium Entretiens Jacques Cartier Information and Communications Technologies: Are they Green?, Montreal, Oct. 2011.

Publication


M. Butt, O. Delgado, and M. Coates, ”An Energy-efficiency Assessment of Content Centric Networking (CCN),” IEEE 25th Annual Canadian Conference on Electrical and Computer Engineering, Mar. 2012.

Additional Resources


References

[1] V. Jacobson, D. Smetters, J. Thornton, M. Plass, N. Briggs, and R. Braynard, “Networking named content,” in Proc. Int. Conf. Emerging Networking Experiments and Tech. (CoNEXT), Rome, Italy, Dec. 2009.

[2] B. Raghavan and J. Ma, “The Energy and Emergy of the Internet,” in Proc. ACM Workshop Hot Topics in Networks, Cambridge, MA, USA, Nov. 2011.

[3] G. Carofiglio, M. Gallo, L. Muscariello, and D. Perino, “Modeling data transfer in content-centric networking,” in Proc. Int. Teletraffic Congress, San- Francisco, California,USA, Sept. 2011.

[4] A. Seetharam, M. Somasundaram, D. Towsley, J. Kurose, and P. Shenoy, “Shipping to streaming: is this shift green?” in Proc. ACM SIGCOMM Workshop Green Networking, New Delhi, India, Aug. 2010.