Internet-Scale High-Level Simulation of the Global Rendezvous Overlay of the PURSUIT Information-Centric Network Architecture

Abstract:

The deployment and operation of a global network architecture can show complex behaviors and the comprehensive validation of these aspects, without building and running the actual system, can only be achieved via simulation. Packet level simulations can be used to test small scale installations, but these experiments may not reveal emerging phenomena stemming from the scaling of the system to its intended real-world size or unnoticed properties of the actual environment, that are not captured by the smaller scale simulation.

We have utilized a high-level model for the validation of the PSIRP/PURSUIT global rendezvous system [1], which comprises of multiple rendezvous networks joined using a hierarchical DHT. The network model is based on the CAIDA AS-level relationship dataset combined with a generated delay space. The traffic matrix used is derived from the three-dimensional utility model for ASes [2] and the rough assumption of Zipfian popularity distribution of content items [3,4,etc.]. We have developed a simple evolution model for the deployment of rendezvous nodes based on the AS business relationships and simplified the rendezvous node operation with an analytical model for caching and using a linear model for the rendezvous operations by assuming that they do not interact. Our initial results include the estimate for the number of needed nodes and their locations and the effect of the cache size parameter to the latency and stretch distributions for individual rendezvous operations showing the scalability and efficiency of the system.

We are currently rewriting the existing single-node Python implementation into a general purpose parallel discrete event simulator and improving the network and traffic models used. This allows us to simulate the rendezvous nodes in full detail and study dependencies in the traffic patterns such as flash crowds. We will also model a DDoS attack against the system, network failures, the total energy consumption of the system, and try to add POP-level detail to the network model.


[1] J. Rajahalme, M. Särelä, K. Visala, and J. Riihijärvi. On name-based inter-domain routing. In Computer Networks Journal: Special Issue on Architectures and Protocols for the Future Internet, 55, 975-986, 2011

[2] H. Chang, S. Jamin, Z. Morley, and M. W. Willinger. An Empirical Approach to Modeling Inter-AS Traffic Matrices. In ACM SIGCOMM IMC’07. Proceedings, pages 139-152, 2005

[3] M. Cha, H. Kwak, P. Rodriguez, Y. Ahn, and S. Moon. I Tube, You Tube, Everybody Tubes: Analyzing the World’s Largest User Generated Content Video System. In ACM SIGCOMM IMC’07. Proceedings, pages 1-14, 2007

[4] P. Gill, M. Arlitt, Z. Li, and A Mahanti. Youtube Traffic Characterization: A View From the Edge. In ACM SIGCOMM IMC’07. Proceedings, pages, 15-28, 2007


Bio:

Kari Visala is a researcher at HIIT working in the PURSUIT EU project and a PhD student at Aalto University. He is currently focusing on the inter-domain issues, evaluation, and the core specification of the clean-slate, information-centric network architecture developed in the project. He received his M.Sc. in Computer Science from Tampere University of Technology and has previously been working in industry, in places ranging from Nokia Research Center to the Mediaclick Internet startup, of which he was one of the founders. Kari has won a silver medal in International Olympiad in Informatics, is a European 4 dan go player, and ran his first marathon in 2011.