Hi guys,
I would like to know somethimg more specific for
indexes and clusters as far as I do not know very much
of R-TREES.
I have this situation:
create table gsc_act(
reg_id int2,
id int2,
ra int4,
de int4,
mag int2
);
This is a huge table approx 2GB containig data for
about 25 milion stars on the sky. On insertion table
was sorted by "de";
I created two indexes:
create index gsc_rdm on gsc_act(ra,de,mag);
create index gsc_dm on gsc_act(de,mag);
queryes using this indexes worked fine. But i decided
to make:
cluster gsc_rdm on gsc_act;
The result was Extremely fast queryes using gsc_rdm
index and gsc_dm index dissapeared. I recreated it but
it was very slow.
So will it be better to create the first index this
way (and is there any difference) as far as the table
is sorted by "de" on inserion:
create index gsc_dmr on gsc_act(de,mag,ra);
or
create index gsc_drm on gsc_act(de,ra,mag);
Where can i read more about how exactly indexes are
created (programmers point of view) online?
Thanks in advance! Waiting for your help!
regards,
Rumen
__________________________________________________
Get personalized email addresses from Yahoo! Mail - only $35
a year! http://personal.mail.yahoo.com/
