Browse Prior Art Database

Estimation of Excessive/Acceptable Channel Overruns

IP.com Disclosure Number: IPCOM000087404D
Original Publication Date: 1977-Jan-01
Included in the Prior Art Database: 2005-Mar-03
Document File: 2 page(s) / 64K

Publishing Venue

IBM

Related People

Dellheim, JL: AUTHOR

Abstract

"Excessive" channel overruns for various configurations of electronic computing systems can be simply "forecast" and visualized by moderately skilled persons.

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 53% of the total text.

Page 1 of 2

Estimation of Excessive/Acceptable Channel Overruns

"Excessive" channel overruns for various configurations of electronic computing systems can be simply "forecast" and visualized by moderately skilled persons.

Channel overrun refers to a condition in which an input-output channel cannot complete a transfer of information in a required finite interval of time. It can be caused by contending processing functions and/or cabling delays. Channel overrun inhibits proper operation of a computing system and tends to degrade system performance. "Excessive overrun", as used herein, means a frequency of overrun recurrence which is considered commercially unacceptable.

The graphic charting and display technique described herein provides a convenient and effective tool for forecasting and visualizing excessiveness or acceptability of channel overrun events for a variety of system configurations without complex mathematics or calculations. The basic factors are: 1. A "channel starting point" table based on empirically developed statistics. 2. A "logical storage unit" (LSU) use table. 3. A "datum time selection" table based on empirical statistics. 4. A means of graphic display (graph paper, CRT, etc.) for comparative indication of three ordinate parameters per channel in a system configuration (namely, "data needed", "service resolutions" and "L:SU usage" on a conmon abscissa scale representing system operating time.

Referring to the accompanying illustration the procedure for determining statistical acceptability of overrun frequency operates as follows: 1. The channels of the configuration to be "tested" (for instance twelve channels) are assigned priorities (for instance, in the range one to seven and nine to thirteen).
2. The channel and priority are used to extract a respective starting-point value fro...