A case for context-aware TCP/IP

Publisher Website
Google Scholar
Abstract
This paper discusses the design and evaluation of CATNIP, a Context-Aware Transport/Network Internet Protocol for the Web. This integrated protocol uses application-layer knowledge (i.e., Web document size) to provide explicit context information to the TCP and IP protocols. While this approach violates the traditional layered Internet protocol architecture, it enables informed decision-making, both at network endpoints and at network touters, regarding flow control, congestion control, and packet discard decisions.We evaluate the performance of the context-aware TCP/IP approach first using ns-2 network simulation, and then using WAN emulation to test a prototype implementation of CATNIP in the Linux kernel of an Apache Web server. The advantages of the CATNIP approach are particularly evident in a congested Internet with 1-10% packet loss. Simulation results indicate a 10-20% reduction in TCP packet loss using simple endpoint control mechanisms, with no adverse impact on Web page retrieval times. More importantly, using CATNIP context information at IP touters can reduce mean Web page retrieval times by 20-80%, and the standard deviation by 60-90%. The CATNIP algorithm can also interoperate with Random Early Detection (RED) for active queue management.

This publication has 21 references indexed in Scilit: