On Thu, 10 Jan 2008, J?rg Kiegeland wrote:
Did you try integer arrays with GIN (inverted index) ?
I now tried this, and GIN turned out to be linear time, compared with GIST
which was acceptable time. However I tested this only for Z=infinity, for
Z=1000, GIST/GIN are both not acceptable.
Sor
Did you try integer arrays with GIN (inverted index) ?
I now tried this, and GIN turned out to be linear time, compared with
GIST which was acceptable time. However I tested this only for
Z=infinity, for Z=1000, GIST/GIN are both not acceptable.
---(end of broadcast)--
Did you try integer arrays with GIN (inverted index) ?
Oleg
On Wed, 9 Jan 2008, J?rg Kiegeland wrote:
Hello,
I have an interesting generic search task, for which I have done different
performance tests and I would like to share and discuss my results on this
newsgroup.
So I begin to descri
Hello,
I have an interesting generic search task, for which I have done
different performance tests and I would like to share and discuss my
results on this newsgroup.
So I begin to describe the search task:
=
You have a set of N unique IDs. Every ID is associated with an integer
sc
