Dnia 2003-12-22 11:39, Użytkownik Ryszard Lach napisał:
> I have a table with 24k records and btree index on column 'id'. Is this
> normal, that 'select max(id)' or 'select count(id)' causes a sequential
> scan? It takes over 24 seconds (on a pretty fast machine):
Yes, this is normal. Because of MVCC all rows must be checked and
Postgres doesn't cache count(*) like Mysql.
This is also normal, but try to change this query into:
select id from some_table order by id desc limit 1;
What is your Postgresql version?
In response to
pgsql-performance by date
|Next:||From: Michael Guerin||Date: 2003-12-22 19:11:54|
|Subject: postgresql performance on linux port|
|Previous:||From: Pavel Stehule||Date: 2003-12-22 11:03:05|
|Subject: Re: "select max/count(id)" not using index|