repository
컨텐트 중심 네트워킹을 위한 효율적인 라우팅 및 캐시 관리 기법
- Alternative Title
- Sung-Won Lee
- Author(s)
- Lee, Sung Won
- Alternative Author(s)
- Sung-Won Lee
- Advisor
- 고영배
- Department
- 일반대학원 컴퓨터공학과
- Publisher
- The Graduate School, Ajou University
- Publication Year
- 2014-02
- Language
- eng
- Keyword
- 컨텐트 중심 네트워킹; 라우팅; 캐시 관리; Content Centric Networking; Routing; Forwarding; Cache Management
- Alternative Abstract
- Content-Centric Networking (CCN) is a new, redesigned communication networking architecture of the Internet to replace traditional host-centric communication with content-centric communication. CCN proposes an architecture which is centered on the content itself, without regard to where it is physically located. The CCN architecture includes an in-network caching and a content-based routing as main features where every content has its own name as an identifier. Although content-centric networking embraces caching and content-based routing, there is little work on the comprehensive interaction between them. Most ongoing research on content-centric networking is centered on independent research item among emerging contents analysis, content naming, caching technology, and content-oriented routing. Therefore, we have an aim for synergy between cache management and content-based routing. To reliably evaluate CCN performances, we developed OPNET CCN Simulator which is easy to configure diverse network environments. It covers most functionality of existing CCN simulators and overcome their drawbacks by providing GUI-based configuration, heterogeneous cache management and CCN topology generator software that allows users to generate diverse Internet-like topologies with node naming. Then, we propose the lightweight prefix-based routing protocol with a consideration of the complexity of the longest-prefix matching in CCN. The proposed scheme limits the area of route discovery to reduce the number of entries in the routing table and FIB with aggregation mechanism. Furthermore, the LPR provides a method of using inter-hierarchical connections to reduce overall route lengths. As an interactive mechanism between cache replacement and cache-aware routing, the pheromone-based cache replacement and forwarding algorithm is proposed with a new metric, pheromone. This metric represents a degree of content concentration for a particular category. In addition, the metric is utilized to diverse the cached contents in the local area and to forward Interest to a router which has been expected to cache the required content. Then, the forwarding algorithm leads to concentrate contents of the particular category in the router even if cache miss occurs. In addition, the cache capacity-aware content centric networking is proposed for an interaction between cache replication and cache-aware routing. This proposal considers cache capacity as a main metric which is a supportable caching space and can be defined with a willingness degree by the previous proposal. The proposed scheme is composed of a selective caching and a cache-aware routing algorithm which provides deterministic cache-awareness without any control message. Then, with analysis about CCN performance under flash crowd, the CC-CCN mechanism is extended with flash crowd detection and related confrontational procedures to reduce network impact by flash crowd. Simulation studies for each proposed scheme show that content-based routing mechanisms in both sides of server-aware and cache-aware routing need little control overhead and help cache hit ratio to be improved. In addition, a synergy effect between the proposed cache management algorithms and the cache-aware routing makes significant caching performance improvement and network load reduction even in flash crowd effect.
- Fulltext
- Appears in Collections:
- Graduate School of Ajou University > Department of Computer Engineering > 4. Theses(Ph.D)
- Files in This Item:
- There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
-
- License
- STATISTICS
- Total Visit :3,723,471
- Total Download :1,818
- Today View :1,129
