Workflow Inbox Performance Issues

• 
• 

Details

Category: Workflow

Published on Thursday, 23 October 2008 16:51

Written by Anon.

Hits: 9496

The performance in the workflow inbox depends on several factors:

1. Use of organizational objects (O, S ...) for tasks

If agent assignment for tasks is controlled using organizational objects, a structural explosion is necessary. Complex organizational structures can mean that the time required to construct the inbox is greater. This is carried out once a day for each user when the user logs on initially, and the information is then buffered. As a result, the initial construction of the inbox takes the most time.

{googleAds}
<script type="text/javascript"><!--
google_ad_client = "pub-2723892367221695";
/* 728x90, created 10/26/08 */
google_ad_slot = "0440586343";
google_ad_width = 728;
google_ad_height = 90;
//-->
</script>
<script type="text/javascript"
src="http://pagead2.googlesyndication.com/pagead/show_ads.js">
</script>
{/googleAds}

2. Number of work items in the system

The time required for selecting work items for a user increases with the number of work items that exist in the system.

3. Number of work items that are displayed in the workflow inbox or in the subordinate views.

4. Use of dynamic columns

The dynamic columns are filled on a task-specific basis with content that is determined dynamically during runtime. To do this, it is necessary for the system to read the work item container and then to determine the attributes to be displayed, for example, of an business object.
Displaying the dynamic columns greatly affects performance, since the system accesses both the work item container and application tables.

5. Grouping according to contents, content type and sort key

Each work item carries the two container elements _WI_Object_ID (dominant object) and _WI_Group_ID (Grouping characteristic) to its container. Both elements are defined so that an object reference can be included.

Grouping according to content and content type is based on the container element _WI_Object_ID, and grouping according to sort key is based on _WI_Group_ID. For each of these grouping types, the system needs to access the container for each work item to read the object reference. In addition, the system must determine the default attribute of the respective object, which involves accessing the application tables.

6. Showing the columns group object and work item contents

As above, showing the two columns is based on the container elements _WI_Object_ID and _WI_Group_ID. The system determines the values to be shown by reading the work item container and the default attribute of the relevant object.

• < Prev
• Next >