A transport protocol for content-centric networking with explicit congestion control

Abstract

Content-centric networking (CCN) adopts a receiver-driven, hop-by-hop transport approach that facilitates in-network caching, which in turn leads to multiple sources and multiple paths for transferring content. In such a case, keeping a single round trip time (RTT) estimator for a multi-path flow is insufficient as each path may experience different round trip times. To solve this problem, it has been proposed to use multiple RTT estimators to predict network condition. In this paper, we examine an alternative approach to this problem, CHoPCoP, which utilizes explicit congestion control to cope with the multiple-source, multiple-path situation. Protocol design innovations of CHoPCoP include a random early marking (REM) scheme that explicitly signals network congestion, and a per-hop fair share Interest shaping algorithm (FISP) and a receiver Interest control method (RIC) that regulate the Interest rates at routers and the receiver respectively. We have implemented CHoPCoP on the ORBIT testbed and conducted experiments under various network and traffic settings. The evaluation shows that CHoPCoP is a viable approach that can effectively deal with congestion in the multipath environment. I. INTRODUCTION During the last decade, content retrieval has dominated the Internet usage. To address the challenges posed for content retrieval, content-centric networking (CCN) (12), (15) has been proposed. Being a significant shift in the network design philosophy, CCN is centered on named content instead of host addresses. Routing towards a content is based on the content name instead of the host address, and data retrieval is initiated by issuing Interest at the content receiver. Com- pared to application-layer overlay solutions such as Content Distribution Networks (CDN) and Peer-to-peer systems (P2P), CCN holds the promise of providing a more efficient and cost- effective solution to content dissemination. CCN's unique characteristics introduce new design chal- lenges for the underlying transport protocol. First, CCN is naturally receiver-driven, since the content receiver needs to issue an Interest first in order to request a Data chunk. Second, hop-by-hop transfer is desired for CCN transport, because con- tent files can be cached along the route to improve throughput. Moreover, since a specific content is often widely disseminated and cached in the network, a CCN flow may have multiple sources - i.e., one content chunk originates from source A, while the next chunk might originate from source B. Such multi-source/multi-path transfer in CCN makes congestion estimation based on a single RTT value fall short. These features of CCN are sufficiently distinct from a traditional end-to-end host-based model that a new transport approach is called for. Recently transport protocol design for CCN has received