Re: Why frequently updated tables are an issue

Jan Wieck <JanWieck(at)Yahoo(dot)com> writes:
> On 6/12/2004 3:45 PM, Tom Lane wrote:
>> I don't think it would help very much to define a bit like that --- I
>> can't believe that very many pages would contain only frozen tuples,
>> unless you were to adopt an aggressive policy of using VACUUM FREEZE
>> a lot.

> I thought this implies an aggressive policy of freezing everything by
> default. But I guess there is something I am not aware of that makes
> aggressive freezing a bad thing.

Well, it means extra I/O to freeze tuples that you otherwise probably
never would. So it's not obvious that aggressive freezing in hopes of
saving cycles later is a win.

>> It might be interesting though to have some kind of "fast vacuum" mode
>> that doesn't worry about freezing tuples, but only reclaiming dead ones.

> Wouldn't that screw the current FSM population mechanisms? Not that my
> suggestions above wouldn't do that either :-)

Yeah, that's another "wholesale" mechanism that we'd have to look at
refining.

regards, tom lane