Showing papers by "Craig Partridge published in 1998"

Recommendations on Queue Management and Congestion Avoidance in the Internet

Abstract: This memo presents two recommendations to the Internet community concerning measures to improve and preserve Internet performance. It presents a strong recommendation for testing, standardization, and widespread deployment of active queue management in routers, to improve the performance of today's Internet. It also urges a concerted effort of research, measurement, and ultimate deployment of router mechanisms to protect the Internet from flows that are not sufficiently responsive to congestion notification.

Increasing TCP's Initial Window

Abstract: This document specifies an optional standard for TCP to increase the permitted initial window from one or two segment(s) to roughly 4K bytes, replacing RFC 2414. It discusses the advantages and disadvantages of the higher initial window, and includes discussion of experiments and simulations showing that the higher initial window does not lead to congestion collapse. Finally, this document provides guidance on implementation issues.

A 50-Gb/s IP router

Abstract: Aggressive research on gigabit-per-second networks has led to dramatic improvements in network transmission speeds. One result of these improvements has been to put pressure on router technology to keep pace. This paper describes a router, nearly completed, which is more than fast enough to keep up with the latest transmission technologies. The router has a backplane speed of 50 Gb/s and can forward tens of millions of packets per second.

Performance of checksums and CRC's over real data

Abstract: Checksum and cyclic redundancy check (CRC) algorithms have historically been studied under the assumption that the data fed to the algorithms was uniformly distributed This paper examines the behavior of checksums and CRCs over real data from various UNIX file systems We show that, when given real data in small to modest pieces (eg, 48 bytes), all the checksum algorithms have skewed distributions These results have implications for CRCs and checksums when applied to real data They also can cause a spectacular failure rate for both the TCP and ones-complement Fletcher (1983) checksums when trying to detect certain types of packet splices When measured over several large file systems, the 16 bit TCP checksum performed about as well as a 10-bit CRC We show that for fragmentation-and-reassembly error models, the checksum contribution of each fragment are, in effect, colored by the fragment's offset in the splice This coloring explains the performance of Fletcher's sum on nonuniform data, and shows that placing checksum fields in a packet trailer is theoretically no worse than a header checksum field In practice, the TCP trailer sums outperform even Fletcher header sums

Method and apparatus for multiplexing bytes over parallel communications links using data slices

Abstract: A communication technique for transmitting packet data over parallel communication sublinks coupled to a processor unit is provided. Initially, a method receives a packet of data from a first communication link which is coupled to a set of sublinks. The method distributes packets over each of the sublinks utilizing a unique byte-by-byte (BBB) striping technique. Logically, the data bytes associated with each sublink are collected into a slice of data and each set of slices are given a unique predetermined label. Each slice is then synchronously transmitted at the aggregate bandwidth of each sublink in parallel across each corresponding sublink to a receiver. A receiver receives the slices of data from the set of sublinks and aggregates the bandwidth of these two or more communication sublinks into a single communication link. Unless there are errors, a packet is transmitted in order using multiple slices. The system recreates the original packet of data from sets of slices having the same unique label. Specifically, the system uses the byte-by-byte striping technique to extract the appropriate bytes of information from each slice received over the parallel sublinks based upon a predetermined sublink sequence corresponding to the labels. This technique is advantageous in that it provides an optimal balance between preserving packet order and conserving network resources.