Page 1 of 1

File matching using splice and performance

PostPosted: Sun Mar 21, 2010 10:55 pm
by pacha
I am using ICETOOL SPLICE operator to delete records from file A, if it matches certain fields of records in file B.

File A

EMPNO NAME COUNTRY
111 RAJU INDIA
222 SHYAM JAPAN
333 RAM INDIA
444 SWA CHINA


FILE B
TOTAL-EMP COUNTRY
50 INDIA
60 CHINA


Records in file A should be delted if country matches in both files. I was able to resolve this using SPLICE, KEEPNODUPES and KEEBASE. But the problem is my file A is having millions of records. But even if my file B is empty or having few records, splice takes a lot of time to execute. Please help me out to resolve this.

Re: File matching using splice and performance

PostPosted: Tue Mar 23, 2010 10:53 pm
by Frank Yaeger
"A lot of time" is rather vague, but you could try using JOINKEYS instead of SPLICE. For complete information on DFSORT's JOINKEYS function, see:

http://www.ibm.com/support/docview.wss? ... g3T7000174