Re: Fillfactor for GIN indexes

On Thu, Jan 15, 2015 at 7:06 AM, Michael Paquier
<michael(dot)paquier(at)gmail(dot)com> wrote:
> Alexander Korotkov wrote:
>> I'm not sure. On the one hand it's unclear why fillfactor should be
>> different from 9.4.
>> On the other hand it's unclear why it should be different from btree.
>> I propose marking this "ready for committer". So, committer can make a final
>> decision.
> OK let's do so then. My preference is to fully pack the index at
> build. GIN compression has been one of the headlines of 9.4.

I'm struggling to understand why we shouldn't just reject this patch.
On November 27th, Cedric said:

"what are the benefits of this patch ? (maybe you had some test case
or a benchmark ?)"

Nobody replied. On January 15th, you (Michael) hypothesized that
"this patch has value to control random updates on GIN indexes" but
there seem to be absolutely no test results showing that any such
value exists.

There's only value in adding a fillfactor parameter to GIN indexes if
it improves performance. There are no benchmarks showing it does.
So, why are we still talking about this?

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