Browse Prior Art Database

Bandwidth Usage in Peer-To-Peer Closed Rings

IP.com Disclosure Number: IPCOM000046990D
Original Publication Date: 1983-Sep-01
Included in the Prior Art Database: 2005-Feb-07
Document File: 4 page(s) / 73K

Publishing Venue

IBM

Related People

Lanier, CS: AUTHOR [+4]

Abstract

In certain distributed data processing systems, data processing stations communicate as peers in closed-ring network structures. In such networks, "informational" frames containing origin and destination address indications and varying amounts of data or control information are circulated from station to station along the ring, and eventually are removed from the ring at either a destination station or at the origin station. Received informational frames are acknowledged by minimal length response frames before the sending station sends additional informational frames. An important objective in such systems is to secure removal of information in the informational frames at destination stations even if a receiving station cannot process the information (e.g.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 28% of the total text.

Page 1 of 4

Bandwidth Usage in Peer-To-Peer Closed Rings

In certain distributed data processing systems, data processing stations communicate as peers in closed-ring network structures. In such networks, "informational" frames containing origin and destination address indications and varying amounts of data or control information are circulated from station to station along the ring, and eventually are removed from the ring at either a destination station or at the origin station. Received informational frames are acknowledged by minimal length response frames before the sending station sends additional informational frames. An important objective in such systems is to secure removal of information in the informational frames at destination stations even if a receiving station cannot process the information (e.g., because of preoccupation with other processes or lack of authorization or other circumstance). Such "early" removal makes improved ring bandwidth usage. Protocols described herein are designed to meet this objective. Fig. 1 exemplifies the subject ring network structure with only four stations shown, i.e., stations A, B, C and D (although it will be understood that many additional stations could participate). The stations are basically identical and communicate as peers. Each station contains a host data processing system, and a microprocessor- controlled ring communication controller (RCC) which interfaces between the ring medium R and the respective host. The details of station structure indicated at station A apply to all stations. As shown at station A, each station has a receiving port 1, a sending port 2 and RCC circuits 3 interfacing between the sending and receiving ports. A bypass relay (not shown) at each station allows the station to directly connect its receiving port to its sending port for error recovery purposes, which is not relevant to the present discussion. The RCC equipment comprises circuits 6 for receiving and demodulating digital signals presented at the receiving port, circuits 7 for transmitting digital signals to the sending port in modulated form suitable for transmission on the ring, circuits 8 for selectively routing demodulated signals either internally or to circuits interfacing with circuits 7, circuits 9 for selectively applying local origin information and through traffic to circuits 7, and insertion buffer 10 (also termed a front end queue (FEQ)) for storing through traffic signals passed along by routing circuits 8 and for applying such signals after varying delays to source selection circuits 9. The RCC equipment also includes an input buffer (IB) 11 for storing received signals having local destinations, which are presented by routing circuits 8, for delayed presentation to the respective host system, an output buffer (OB) 12 for storing locally originated information in transit between the respective host system and source selection (SS) circuits 9, and a microprocessor 13 for controlling and o...