Showing papers by "Craig Partridge published in 1997"
01 Sep 1997
TL;DR: This memo describes the network element behavior required to deliver a guaranteed service (guaranteed delay and bandwidth) in the Internet and follows the service specification template described in [1].
Abstract: This memo describes the network element behavior required to deliver a guaranteed service (guaranteed delay and bandwidth) in the Internet Guaranteed service provides firm (mathematically provable) bounds on end-to-end datagram queueing delays This service makes it possible to provide a service that guarantees both delay and bandwidth This specification follows the service specification template described in [1]
1,115 citations
TL;DR: Some of the reasons TCP/IP has difficulty with satellite links are explained, solutions to some problems are presented, and the state of the research on some the unsolved problems are described.
Abstract: Achieving high data rates using TCP/IP over satellite networks can be difficult. This article explains some of the reasons TCP/IP has difficulty with satellite links. We present solutions to some problems, and describe the state of the research on some the unsolved problems.
361 citations
01 Dec 1997
TL;DR: The designers of the TCP/IP protocol suite explicitly included support of satellites in their design goals, and it has been demonstrated to work over many differing networking technologies, including over paths including satellites links.
Abstract: The designers of the TCP/IP protocol suite explicitly included support of satellites in their design goals. The goal of the Internet Project was to design a protocol which could be layered over different networking technologies to allow them to be concatenated into an internet. The results of this project included two protocols, IP and TCP. IP is the protocol used by all elements in the network and it defines the standard packet format for IP datagrams. TCP is the end-to-end transport protocol commonly used between end systems on the Internet to derive a reliable bi-directional byte-pipe service from the underlying unreliable IP datagram service. Satellite links are explicitly mentioned in Vint Cerf's 2-page article which appeared in 1980 in CCR [2] to introduce the specifications for IP and TCP. In the past fifteen years, TCP has been demonstrated to work over many differing networking technologies, including over paths including satellites links. So if satellite links were in the minds of the designers from the beginning, what is the problem? The problem is that the performance of TCP has in some cases been disappointing. A goal of the authors of the original specification of TCP was to specify only enough behavior to ensure interoperability. The specification left a number of important decisions, in particular how much data is to be sent when, to the implementor. This was deliberately' done. By leaving performance-related decisions to the implementor, this would allow the protocol TCP to be tuned and adapted to different networks and situations in the future without the need to revise the specification of the protocol, or break interoperability. Interoperability would continue while future implementations would be allowed flexibility to adapt to needs which could not be anticipated at the time of the original protocol design.