And again...read the doc's linked to previousily...
From the REDBOOK - Striving for Optimal Journal Performance on DB2
Universal Database for iSeries sg246286..
I've read this about 10 years ago and besides it's outdated, nonsens is getting any better, if you rewrite it again and again and recommendations to put commitment controll to existing batch jobs by nesting them with a STRCMTCTL and commit and ENDCMTCTL are not only nonsense, this is dangerous!!!
I'm tired and I will go to bed, have a nice weekend.