Browse Prior Art Database

Automated Siebel application usage analysis

IP.com Disclosure Number: IPCOM000195117D
Publication Date: 2010-Apr-21
Document File: 5 page(s) / 88K

Publishing Venue

The IP.com Prior Art Database

Abstract

Siebel applications developed over years tend to grow generally quite uncontrolled in terms of application objects (e.g. Applets, Views, Screens, Business Components, eScripts etc.) that built-up the core application. Symptoms amongst others are a huge amount of applets partially not used anymore, complex business services that nobody understands or even worse know whether they are still required or not. These symptoms do have straight influence on maintainability and upgradeability. In some areas performance and system stability can be negatively affected as well. "Automated Siebel application usage analysis" allows to address this issue effectively.

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 45% of the total text.

Page 1 of 5

Automated Siebel application usage analysis

Siebel applications developed over years tend to grow generally quite uncontrolled in terms of application objects (e.g. Applets, Views, Screens, Business Components, eScripts etc.) that built-up the core application. Symptoms amongst others are a huge amount of applets partially not used anymore, complex business services that nobody understands or even worse know whether they are still required or not. These symptoms do have straight influence on maintainability and upgradeability. In some areas performance and system stability can be negatively affected as well.

In most instances nobody kicks-off a project to optimize the application unless one of the areas is clearly and noticeable impacted. A Siebel upgrade project however offers the unique possibility to work and address this area without having to setup a own project that takes care of these issues.

In a classical upgrade project the end users would now be interviewed by business analysts to get an understanding of the GUI parts that are still required. For all functionality behind the GUI a developer would be asked to analyze and identify the object candidates for deactivation. Both actions tend to be very error prone and are ineffective in its result.

The idea of "Automated Siebel application usage analysis" is to leverage the information collected in object manager log files and contained in the repository. With both sources the application usage can be effectively determined.

How does the idea work ?

Step one

During a user session Siebel can output information of views visited into its object manager log file. When Siebel is run with its thin client the log files can be collected at a central directory at the server. When Siebel is run as a dedicated client the object manager log file will be available at the agents PC where it needs to be somehow collected and copied to a central directory.

Once all object manager log files are available the first step of the analysis can be started. The first step examines the object manager log files and extracts all information of "views visited" file by file. This run returns an overview of Siebel views that were visited during all user sessions. For this step and the result it is advantageous to collect log files over several days and not just some hours. This will produce a good statistical result and increases the probability that less frequent used Siebel views are visited at least once.

The result of this is loaded into a (local) database. In the example of this article the results are loaded into TABLE

1:

_

TABLE

Views visited

My Activities

My Activities

_1

1

Page 2 of 5

My Activities

My Service Requests My Activities

......

Still in step one subsequent code is executed that sums-up the views visited and writes the result in TABLE

_

2:

TABLE

_2

Views Times visited

My Activities 34 My Service
Requests

1

......

Visualization of "Step one":

Fig. 1 - Visualization of "step one"

Step two

...