Re: Single client performance on trivial SELECTs

On Thu, Apr 14, 2011 at 12:45 PM, Noah Misch <noah@leadboat.com> wrote:
>> It shouldn't be
>> terribly difficult to come up with some kind of hash function based
>> on, say, the first two characters of the keyword that would be a lot
>> faster than what we're doing now.
>
> I'd look at `gperf', which generates code for this from your keyword list.

I thought of that, but wasn't sure we wanted to introduce a dependency
on that tool.  That might be a pain, especially on Windows.  But maybe
we could steal the basic approach.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company