Re: MIDRANGE-L Digest, Vol 14, Issue 136

I understand the performance penalty, but
I guess somehow bundling journal writes (not making them when they happen)
is different from not writing data changes when they happen?
Still baffles me. I guess IBM just keeps it straight. The more it's like
magic........
Jim

message: 1
date: Thu, 29 Jan 2015 12:02:27 -0500
from: DrFranken <midrange@xxxxxxxxxxxx>
subject: Re: Power HA and Journaling.

Yes force ration of 1 would be bad.

Because journaling bundles writes is the biggest piece. You may have an
interactive program that updates 50 or more rows across a dozen tables.
If you did a force of 1 to every table that means 50 hard writes. If you
journal then you get all 50 of those writes in one 'blob' through the
journal into the receiver. Much like 1 truck with a pallet full of boxes
vs 50 YuGos with one box each. Once they are in there DB management
considers them 'Written' and from there assures they get to the
appropriate tables. This happens asynchronously and after your program
has already moved on.

This is a big part of why turning on journaling for a thousand tables
might not even be noticeable performance wise and in some situations we
have seen Improvements in performance!! Now if you were silly enough to
journal every table to it's own journal, um, we're back to 'bad' :-)

Please ladies and gentlemen, do not fear the journal!!

- Larry "DrFranken" Bolhuis