There have been some performance improvements in 15.1, 15.2, and a defect fixed in 15.3 that have improved performance. Below is a list of the improvements included:
Performance Improvement for Incremental Updates. Performance has been improved for the Incremental Data Warehouse load specifically in the following areas:
- Resource availability
- Project teams
- Task assignments
We have reduced the incremental load time by adding some indexes and using temporary processing tables.
The following performance improvements are also available:
- Financial plans for incremental loads: The incremental load makes use of a new DWH_TMP_FIN_RECORD_KEYS table that contains only those plans that changed since the last incremental load. The database view joins to this table to reduce the amount of data processed. This can significantly improve performance on large data sets.
- Financial transaction facts: Similar to recent improvements for assignments and teams, the load performance of financial transaction facts data is improved. These tables often contain millions of rows of data. The tables include DWH_FIN_TRANSACTION, DWH_FIN_TRANSACTION_LN, and DWH_INV_TASK_PERIOD_FACTS.
Additionally, in 15.3:
The following defect has been fixed that can cause the job to run longer than expected in Incremental mode: KB000115923
Workarounds:
Some suggestions that may help with performance until upgrading the latest CA PPM version:
- Run the Load DWH job in full mode only
- Reduce the amount of custom attributes included in DWH
- Reduce the amount of languages included
- Ensure you are on latest fix pack (patch) for your release
- Make sure that both Time slicing and Investment Allocation are incompatible with the Load DWH job (this can avoid Load DWH job failures in particular): KB000036882