| From: | Milan Oparnica <milan(dot)opa(at)gmail(dot)com> |
|---|---|
| To: | pgsql-sql(at)postgresql(dot)org |
| Subject: | Re: PERSISTANT PREPARE (another point of view) |
| Date: | 2008-07-21 20:45:36 |
| Message-ID: | g62sgn$9b6$1@news.hub.org |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-sql |
I found this link from IBM DB2 developers showing why PERSISTENT PREPARE
is a good idea and how could it be implemented.
http://www.hoadb2ug.org/Docs/Favero20606.pdf
It seems that main benefit (beside efficiency) is memory.
Having number of connections all with dozens of PREPARED statements
consumes lot of memory.
If we put these statements in global space (perhaps not even in RAM
memory until needed) we could reuse optimization plans as well as
physical space needed to store them.
I also found articles:
http://archives.postgresql.org/pgsql-hackers/2008-04/msg00867.php
and
http://archives.postgresql.org/pgsql-performance/2008-03/msg00480.php
all about users trying to explain PERSISTENT PREPARE (some refer to it
as global prepare).
There are also some guys who actually made some code for PERSISTENT PREPARE:
http://archives.postgresql.org/pgsql-hackers/2008-03/msg01228.php
and
http://archives.postgresql.org/pgsql-hackers/2008-03/msg01219.php
PEOPLE NEED THIS FEATURE !!! It is not discovering the wheel but it will
simplify DB programming and even gain some performance.
NONE OF POPULAR SQL DBMS (Oracle, MS SQL, MySQL, Postgre, INTERBASE,
FIREBIRD) HAVE THIS FEATURE.
WHY ?
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Craig Ringer | 2008-07-22 01:04:41 | Re: PERSISTANT PREPARE (another point of view) |
| Previous Message | Emi Lu | 2008-07-21 16:59:02 | Re: Size or efficiency differences "varchar(128) vs. varchar(32)" |