Re: Bug: Buffer cache is not scan resistant

Jeff Davis <pgsql@j-davis.com> writes:
> On Mon, 2007-03-05 at 15:30 -0500, Tom Lane wrote:
>> Strikes me that expressing that parameter as a percentage of
>> shared_buffers might make it less in need of manual tuning ...

> The original patch was a percentage of effective_cache_size, because in
> theory it may be helpful to have this parameter larger than shared
> buffers. Synchronized Scannning can take advantage of OS buffer cache as
> well.

I didn't say you couldn't allow it to be more than 100% ;-).  But basing
it on effective_cache_size strikes me as a bad idea because that parameter
is seldom better than a wild guess.  shared_buffers at least means
something.
        regards, tom lane