Browse Prior Art Database

Correlate important system activity with process state in order to calculate performance metrics

IP.com Disclosure Number: IPCOM000015891D
Original Publication Date: 2002-Apr-07
Included in the Prior Art Database: 2003-Jun-21
Document File: 2 page(s) / 41K

Publishing Venue

IBM

Abstract

Disclosed is a method to gather usage, delay and workflow like performance metrics used in workload management in operating systems without such metrics (like Windows, Linux, etc.) without modifying the kernel of the operating system.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Page 1 of 2

  Correlate important system activity with process state in order to calculate performance metrics

    Disclosed is a method to gather usage, delay and workflow like performance metrics used in workload management in operating systems without such metrics (like Windows, Linux, etc.) without modifying the kernel of the operating system.

   There are three different states for a process in an operating system: * Idle, the process is not using or waiting for resources, for example if the process waits for user input. * Using, the process uses a resource, for example the CPU or an I/O device. * Delayed, the process waits for a resource which is currently used by another process.

   Measuring these three states allows the end user to observe the contention situation of operating systems and workload management algorithms to adjust resource assignments for the processes.

   To obtain state information the kernel or nucleus of an operating system can be modified to deliver the state information from internal control structures to the outside. But such kernel modifications are not always feasible, i.e. for commercial products, kernel changes are impossible.

   This invention uses a different approach which stays outside of the operating system by observing the activity of the operating system kernel. The approach is applicable to any UNIX like operating system as well as to other platforms. As an example this technique is implemented for the Linux operating system and used to observe syst...