Re: LOG_CEHCKPOINT_INTERVAL AND RECOVERY

From: keith boulton <boulke_at_globalnet.co.uk>
Date: Sat, 27 Mar 1999 15:35:25 GMT
Message-ID: <36fcd9a0.11932768@195.147.246.90>

>Lets say you didn't checkpoint frequently, say you have 500meg logs. As we were
>putting the 499'th meg of stuff in the redo log file the system crashed. To
>startup, we have to read and apply 499meg of log.
>

I just tried an experiment with 100M log files and database restart after shutdown abort with no outstanding transactions and the log being about 3/4 full still took well under 1 minute, which seems a price well worth paying for the sort of performance improvement large log files give for batch processes.

My experience in the past is that the biggest problem on restart is the rollback of large outstanding transactions.

Whenever I've asked the question about checkpointing people always seem to parrot some response about a checkpoint every 20 minutes - do you know the origin of this figure? Received on Sat Mar 27 1999 - 09:35:25 CST