EAR 9.2: A Performance Degradation Has Been Detected At Payment Predictor Application Engine Program (AR_PREDICT)
Last updated on FEBRUARY 27, 2018
Applies to:PeopleSoft Enterprise FIN Receivables - Version 9.2 to 9.2 [Release 9]
Information in this document applies to any platform.
The delivered Receivables Payment Predictor Application Engine Program (AR_PREDICT) suffers from a performance degradation on several steps here below listed:
- Application Engine Program, Section, and Step: AR_PREDICTOR2.DEFCUST.CLEANTMP.S 3610 0.020 3610 1384.400 0 0.000 1384.420
- Application Engine Program, Section, and Step: AR_PREDICTOR2.DEFCUST.FILLBU.S 3610 0.100 3610 3367.160 0 0.000 3367.260
- Application Engine Program, Section, and Step: AR_PREDICTOR2.DEFCUST.FILLCUST.D 3610 0.040 3610 1010.970 3610 0.000 1011.010
- Application Engine Program, Section, and Step: AR_PREDICTOR2.PGEN.UPDDSTI2.S 24141 3.300 24141 5727.140 0 0.000 5730.440
- Application Engine Program, Section, and Step: AR_PREDICTOR2.PUPD_WS.UPD_PTIM.S 1 0.000 1666 28389.720 0 0.000 28389.720
- Application Engine Program, Section, and Step: AR_PREDICTOR2.UPDTLTLR.BU_PYMT.S 3610 0.140 3610 1910.880 0 0.000 1911.020
- Application Engine Program, Section, and Step: AR_PREDICTOR2.UPDTLTLR.CUST_PMT.S 3610 0.150 3610 1753.690 0 0.000 1753.840
The problem is not a system matter, but a coding issue (AR_PREDICTOR2) SQL statement(s). Some of the SQL coding forces the system to do full table scans, which really slows down processing, (specially since there are over 1M Deposits tied to roughly 4M Payments).
- Log into the FSCM Online Application as User ID VP1
- Navigate to: Accounts Receivables > Receivables Update > Request Receivables Update
- Define a Run Control ID, and launch AR_UPDATE AE Program
- Among all the sub-programs being called, AE Program ARPREDCT takes a long time to complete
The severe performance issues detected are delaying further transaction processing, specially i the matching of Items with Payments, and posting the Accounting Entries needed.
Changes in Application Engine structural features, more efficient SQL, and some strategically defined indexes should significantly improve the performance. The Payment Predictor process should be able to handle large volume of transaction data in a timely manner.
Sign In with your My Oracle Support account
Don't have a My Oracle Support account? Click to get started
Million Knowledge Articles and hundreds of Community platforms