Re: [HACKERS] database size - Mailing list pgsql-hackers

From Bruce Momjian
Subject Re: [HACKERS] database size
Date
Msg-id 199801071718.MAA10891@candle.pha.pa.us
Whole thread Raw
In response to Re: [HACKERS] database size  (darrenk@insightdist.com (Darren King))
List pgsql-hackers
> I've seen this for Oracle, but I _can't_ find it right now.  I'll dig it
> up tonite...this is driving me nuts trying to remember where it is now.
>
> But this I do have handy!  It's an HTML page from IBM DB2 docs.  A touch
> long, but I found it to most interesting.
>
> If there are any of the linked pages that someone else is interested in,
> contact me and if I have it, I can send it to you off-list.

Interesting that they have "tombstone" records, which sounds like our
time travel that vacuum cleans up.

They recommend (rowsize+8) * 1.5.

Sounds like we are not too bad.

I assume our index overhead is not as large as data rows, but still
significant.  I am adding a mention of it to the FAQ.  That comes up
often too.

    Indexes do not contain the same overhead, but do contain the
    data that is being indexed, so they can be large also.

--
Bruce Momjian
maillist@candle.pha.pa.us

pgsql-hackers by date:

Previous
From: Keith Parks
Date:
Subject: Floating point exceptions.
Next
From: Bruce Momjian
Date:
Subject: varchar/char size