Browse Prior Art Database

Rapid testing by controlling the process time for each task running on distribued-systems

IP.com Disclosure Number: IPCOM000011915D
Original Publication Date: 2003-Mar-26
Included in the Prior Art Database: 2003-Mar-26
Document File: 3 page(s) / 159K

Publishing Venue

IBM

Abstract

A program is disclosed that rapid system testing can be done by controlling the process time for each task running on distributed-systems. To accomplish this, elimination of idle time and adjustment of time processing are performed.

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 56% of the total text.

Page 1 of 3

THIS COPY WAS MADE FROM AN INTERNAL IBM DOCUMENT AND NOT FROM THE PUBLISHED BOOK

JP820021005 Koichiro Kato/Japan/IBM Tsunefumi Watanabe, Hidekazu Fujiwara

  Rapid testing by controlling the process time for each task running on distribued-systems

   Rapid system testing can be done by controlling the process time for each task running on distributed-systems. To accomplish this, elimination of idle time and adjustment of time processing are performed.

Problem:

Recently, data warehouse is one of the major configurations in large distributed systems. In these systems, first of all, data periodically come from several databases which act as data sources. These are then merged into a global data warehouse. This process is sometimes performed through remote connection. Then, the data stored in this global data warehouse are transferred periodically to the local data warehouse (data mart) by data extraction. Finally, the user queries/analyzes the data in the local data warehouse either periodically or in a specific period of time. Normally, to get the results of these analyses/queries, it will require a considerable amount of data. Thus, it will take a long period of time to validate the whole system process.

Global Data

Warehouse

Data warehouse

Data Sources

Utilization

Departments / End Users

 Local Data warehouse

Data Mart

   Data Marge Periodically

Remote Communication

Corporate Wide

Analysis/Query

   Periodically/ In Specific Time

BASE Database

(Head Quarter)

 BASE database (Branch Office 1)

 BASE database (Branch Office 2)

   Periodical Data Extract

Solution

To shorten the development/deployment time, the system clock may be set ahead. Thus, the processes are done in a shorter span of time.

This disclosed would be applied to the following systems:

- Time driven system, not an event driven system - Not a real time system

Scheduling the pseudo standard time

1. For each task's starting time, XFER_START(n), set the START(n) time as: START(n) = XFER_START(n) - TimeLag(n) where TimeLag(n) is the time lag between the time when the Time server submits the "set-time" r...