Posted on January 11th, 2008 by Sanjit Anand ||
Email This Post
| Have you tried OracleappsHub in ipad/iphone/smart Phone? Don't wait. try it today |
It was one of the requirements of my current client, who has replaced SAP to Oracle.In migration there are journals from SAP systems are over 45,000+ lines long, so client are bit concern . I took lot of research over internet and metalink,and refer Oracle benchmark data sheet, it was no help, but still manage those activity.Finally , thought to share…
What we can assume we have to reach the maximum as the journal line ID allows 15 characters so 999,999,999,999,999 is theoretical maximum .
UBS have benchmarked the GL system and upload over 1 million lines per hour.
There is a case mention in document , and a benchmark test was conducted to validate overall performance.
- Data Load: Import and post at least 10M postings in less than 1 hour
- Month End: Perform a “period open” in less than 12 hours (transfer 500M balances into the next month)
- Daily Extract: Extract 500M balances in less than 60 minutes and calculate the daily balance sheet of the bank by cost center in less than 30 minutes
And these are the outcome of benchmark set of the EBS tool
- They Imported and posted 11.42M journals per hour
- Period Open on 500M balances in 11 hours and 8 minutes
- Extracted 500M balances (including the merge of 26 dimension values) in 48 minutes
- Generated the daily balance sheet on the base of 500M balances in 23 minutes
Sound great ! what you say.
Suggested Reading
Related Posts
January 23rd, 2008 at 5:34 am
Actually, how realistic is this statement : “They Imported and posted 11.42M journals per hour”.
Is this workable also if budgetary control is turn on ? Dont think so. Current client facing problem in posting where when the batch containing > 200,000 lines, the posting stuck at fund reservation(glfxgb) and couldn’t complete. Metalink has not solve the issue even the SR on sev 1 for 1 month+. Anyone have idea on this ?
February 2nd, 2008 at 8:22 am
as my experience, I import 1.2 million in 2 hours with the suggestion from Note:198437.1