How To Improve the Performance of the Refresh Collection Snapshots When Running Very High Volumes
(Doc ID 550005.1)
Last updated on SEPTEMBER 28, 2018
Applies to:Oracle Advanced Supply Chain Planning - Version 11.5.10 and later
Information in this document applies to any platform.
EXECUTABLE:MSRFWOR - Refresh Collection Snapshots
This note shows very advanced techniques to managing MLOG$ Tables when specific circumstances are encountered with performance of Refresh Collection Snapshots.
You should review <Note 1063953.1> to determine if the steps in that note will improve performance, OR if your very high volume and methods of running data collections point to using this note.
In the section 'Explain the Primary Causes of Issues with MLOG$ Tables - Here is a Summary', we provide examples of when this note would be required.
The intention of this note is to help identify very high volume entities that can cause Refresh Collection Snapshots to perform poorly and provide the steps to evaluate and determine which entities could profit from the steps we devised to improve the performance.
When the Planning Data Collection is run, a number of concurrent programs are launched to collect data from the EBS Source tables (instance) to the APS Destination tables (instance).
The Refresh Collection Snapshots program is responsible for taking 'snapshots' of the current data in the EBS Source system. Many times it is this program that causes the biggest performance impact for the entire Planning Data Collection process.
In a distributed installation where APS applications run on separate instance from the EBS Source transaction instance, this program is run on the EBS Source instance.
When we find that the time between running Data Collections is such that very high volumes are recorded in the MLOG tables used to refresh snapshots, we have developed a strategy to help improve performance.
To view full details, sign in with your My Oracle Support account.
Don't have a My Oracle Support account? Click to get started!
In this Document