Re: dbload vs. load

On 8 Feb 2006 00:58:40 -0800, "christrier" <chrishunnell@gmail.com> wrote: > > > Is bullshit 'ClownSpeak' for RTFM? >I have read the fu*** manual that is why i tried to insert the data >with the load from statement. >In the Manual there is a page with 4 questions about choosing the right >import-utility. I had to answer those questions in that manner that i >had choosing load as my result.... > >> IIRC, each operation (in this case, each INSERT) has to check or validate >> something in the log, and does this in a relatively inefficient manner -- >> traversing a linked list. So the longer the transaction gets, the slower >> each subsequent operation is. When the transaction is short, the cost of >> committing is relatively high, but when the transaction gets longer, it >> can often be substantially better to commit more often (if this is >> possible). > >I guess it is better to commit more often and therefore i have another >question: i have tried many values for the commit-intervall and i never >really found out what the max-value was. I tried with 10.000 and it >didn't workout. I have informix sql version 9.40 You're trying to find the max value for commit interval? Seems that would depend on how much logical log space you have . . . . . JWC

John Carlson said: > > On 8 Feb 2006 00:58:40 -0800, "christrier" <chrishunnell@gmail.com> > wrote: > > > You're trying to find the max value for commit interval? Seems that > would depend on how much logical log space you have . . . . . And what else is going on with your server. -- Bye now, Obnoxio "C'est pas parce qu'on n'a rien ' dire qu'il faut fermer sa gueule" - Coluche did i mention i like nulls? heck, i even go so far as to say that all columns in a table except the primary key could/should be nullable. this has certain advantages, for example, if you need to insert a child record and you don't have a parent row for it, just do an insert into the parent table with the primary key value (everything else null), and voila, relational integrity is preserved. but this is, admittedly, a bit controversial among modellers. --r937, dbforums.com