Resource management and optimization in content oriented networks

PhD Thesis Proposal Defence


Title: "Resource management and optimization in content oriented networks"

by

Mr. Min WANG


Abstract:

Today, the Internet is increasingly a platform of online services and 
massive content diffusion has become a daily routine over the Internet. 
However, the Internet was originally designed to primarily support 
pairwise host-to-host communications and thus is poorly suited for content 
dissemination among multiple hosts. With video content becoming 
increasingly dominant and voluminous, the host-centric paradigm reveals 
its inefficiencies more seriously. In recent years, we have seen a variety 
of attempts to provide more efficient content delivery support for the 
Internet, including i) the research efforts of content-centric networking 
(CCN) as a clean-slate redesign of the Internet architecture; ii) the wide 
adoption of CDN services; and iii) the fast expansion of cloud 
infrastructures offering vital support for modern data-intensive 
applications. We coin all these attempts to make the Internet a better 
platform for content distribution as the "content-oriented networks". In 
this thesis, we address several important resource management and 
optimization problems in those content-oriented networks.

This thesis consists of three parts. The first part is about content 
management in CCN. One of the defining features of CCN is pervasive 
in-network caching. The original CCN proposal adopts the ubiquitous-LRU 
caching scheme, which leads to serious on-path redundancy and poor overall 
caching performance. We begin with proposing a new caching scheme PCP and 
demonstrate via trace-driven simulations that PCP outperforms 
ubiquitous-LRU in both microscopic and macroscopic views. Apart from PCP, 
many other new caching schemes have also been proposed. Most of them 
including PCP focus on reducing on-path cache redundancy by invoking 
either explicit or implicit cooperations between nodes along the response 
path. This general approach inherently suffers from issues of scalability 
and hurdles of uniform enforcement across different ASes. To overcome 
these limitations, we next propose an intra-AS cache cooperation scheme 
iCCS, which commits to eliminating both on-path and off-path cache 
redundancy within an AS via a periodic independent procedure. We formulate 
the cooperative redundancy elimination problem CRE-P and then propose a 
distributed greedy algorithm. Through trace-based simulations on multiple 
realistic network topologies, we show that iCCS significantly shortens the 
request-response latency and dramatically reduces the amount of transit 
traffic without increasing internal link congestions. After that, we study 
the content peering problem in CCN by proposing the content-level peering 
model (CPP). With extensive numerical experiments under realistic AS-level 
peering graphs, we find that interconnectivity of the peering graph 
significantly influences the maximum peering benefit, and that cooperative 
caching yields higher peering benefits than local greedy caching but is 
sensitive to parameters like peering link bandwidth and AS-level cache 
size. The second part is about content multi-homing with multiple CDNs. We 
propose the MCDN-CM model, aiming to make an optimal operating plan for 
the authoritative CDN by deciding how to supply content requests with 
guaranteed QoS. We show via numerical experiments under realistic settings 
that MCDN-CM achieves a tremendous cost-saving for the authoritative CDN 
compared to other approaches. The third part is about traffic management 
in inter-data center network. We propose the MCTEQ model, which adopts 
utility-based joint-bandwidth allocation for multiple classes of traffic 
and in particular provides end-to-end delay guarantee for interactive 
flows. We demonstrate via experiments with two realistic inter-data center 
networks that MCTEQ achieves about 160 Gbps higher network utilization 
than the best existing solutions, while running at least twice faster.


Date:			Tuesday, 24 March 2015

Time:                   10:30am - 12:30pm

Venue:                  Room 3494
                        lifts 25/26

Committee Members:	Dr. Brahim Bensaou (Supervisor)
  			Dr. Pan Hui (Chairperson)
  			Prof. Gary Chan
  			Dr. Jogesh Muppala


**** ALL are Welcome ****