You should look at what MySQL is doing, start by looking at SHOW STATUS and SHOW VARIABLES. The SHOW STATUS will give you an idea
how your indexes are performing, thread management, open file management, etc. You can file a lot of info in there and some
variables you can adjust on the fly. There is a bunch of stuff on the internet on how to "read" the status values and determine
ratios. Also, High Performance MySQL I think is an excellent book that cover a lot of bases.
Based on google results, 50 lakhs is about 5,000,000 (5 million) rows. Depending on setup, it may be a lot. I've got "tables" with
over 250 laks without performance issues. I must clarify that the 250 million row "table" is actually a MERGE table with the
underlying tables each holding abuout 20-30 millions rows. We did start having performance slow down before switching to the merge
format. Now once a month a script runs that creates a new empty table and alters the merge table to include the new empty table
where all new records are entered.
You may or may not be able to use MERGE tables in your case. Records are not deleted from our tables, we just drop the oldest table
after a certain length of time (i.e. 6 months).
So how large can a table be? As large as you want.
----- Original Message -----
From: "Ratheesh K J" <[EMAIL PROTECTED]>
To: <mysql@lists.mysql.com>
Sent: Thursday, August 03, 2006 2:42 AM
Subject: Table analysis - Help required urgently
Hello all,
I am required to analyze all the tables of our system. I need to know the key parameters that should be taken into consideration for
analysis. I am not speaking about ANALYZE TABLE <tbl name>. I am required to manually look into all table structures and pin point
problems ( if any ).
So it would be helpful if i could know about
1) what exactly I have to look for in the tables.
2) Index management. Which fields are to be indexed and which not
3) How big can a table be? We have tables which have more than 50 lakhs of rows. Any select queries, insert queries or update
queries are taking more time to execute. So what is a preferable table size.
4) Any other suggestions
Thanks,
Ratheesh Bhat K J
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
