Browse Prior Art Database

Interactive Response Time Control

IP.com Disclosure Number: IPCOM000053152D
Original Publication Date: 1981-Sep-01
Included in the Prior Art Database: 2005-Feb-12
Document File: 3 page(s) / 27K

Publishing Venue

IBM

Related People

Svoboda, G: AUTHOR

Abstract

This article describes a way to control interactive response times in a time sharing multiple virtual storage system (e.g., TSO response times on OS/VS2 MVS).

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 3

Interactive Response Time Control

This article describes a way to control interactive response times in a time sharing multiple virtual storage system (e.g., TSO response times on OS/VS2 MVS).

Typically, when the workload on a computer system decreases or the capability of the system increases as when a faster processor is installed, the response times seen by the interactive user become shorter. This occurs because the system attempts to always run at peak utilization, making full use of the available computer resources and delivering the best possible response times.

However, the computer installation management may want to limit response time so that interactive users see more consistent response times, even if this causes the system to run at less than peak utilization during periods of light workload. Or, management may want to reserve computer capacity for future applications or for an expected increase in the number of users by limiting response times, thereby isolating the current user from the anticipated changes.

A system parameter can be provided that limits response time so that it is not faster than a specified objective. The response time objective is enforced by delaying an interactive transaction after the command or other input is entered by the user but before any real storage resources are committed to the transaction.

The amount of the delay is calculated at timed intervals so that it is adjusted according to the system's ability to process transactions. The calculation is as follows: delay time = response time objective - average transaction processing time where response time objective is the desired amount of elapsed time before the system responds to a trivial request. (Trivial is defined as requiring less than a specified amount of storage, I/O, and processor resources). and average tr...