Difference between revisions of "Improving Content Manager OnDemand Performance"
Jump to navigation
Jump to search
m (Tweaked Link to Table Segmentation) |
(Many tweaks to the article on IBM CMOD Performance.) |
||
| Line 1: | Line 1: | ||
{{TOCright}} | {{TOCright}} | ||
It's possible to improve IBM CMOD's performance by paying close attention to configuration of Application Groups, Applications, and being careful with the usage of the CMOD utilities. | |||
= Introduction = | = Introduction = | ||
| Line 10: | Line 11: | ||
== Application Group Configuration == | == Application Group Configuration == | ||
* Increase Max # of Rows to reduce [[Table Segmentation in Content Manager OnDemand|Table Segmentation]] in the database | * Increase Max # of Rows to reduce [[Table Segmentation in Content Manager OnDemand|Table Segmentation]] in the database | ||
* Specify the CORRECT Expiration Type (Document, Load, Segment) | * Specify the CORRECT Expiration Type (Document, Load, Segment) | ||
* Use Indexes carefully | * Use Indexes carefully | ||
| Line 20: | Line 21: | ||
== Loading data into Content Manager OnDemand == | == Loading data into Content Manager OnDemand == | ||
* Load in large batches with small date ranges | * Load in large batches (hundreds to thousands of documents at a time) with small date ranges | ||
* Load chronologically | * Load chronologically to prevent breaking table segmentation in CMOD. | ||
* Use separate Load Servers | * Use separate Load Servers | ||
Revision as of 18:00, 10 February 2017
It's possible to improve IBM CMOD's performance by paying close attention to configuration of Application Groups, Applications, and being careful with the usage of the CMOD utilities.
Introduction
When you're first learning about OnDemand, there seems to be a huge focus on making it work by any means possible, without considering the future implications of what you're doing. Given that CMOD is an archive product that can store documents for several years before you begin expiring them, there are a lot of problems that are created in those early days when "get it done" is the priority. This article will explain some of the ways to ensure your OnDemand server keeps its lightning-fast performance.
Architecture of Application Groups and Applications
- Combine business documents into the same Application Group
- Example of Call Center Usage
Application Group Configuration
- Increase Max # of Rows to reduce Table Segmentation in the database
- Specify the CORRECT Expiration Type (Document, Load, Segment)
- Use Indexes carefully
Application Configuration
- Minimize floating triggers
- use TLEs in AFP data
- Use Generic Indexes for PDFs, rather than the PDF indexer
- Use PPDs in PDFs for super-fast indexing in CMOD v9.5 and higher
Loading data into Content Manager OnDemand
- Load in large batches (hundreds to thousands of documents at a time) with small date ranges
- Load chronologically to prevent breaking table segmentation in CMOD.
- Use separate Load Servers