MySQL Forums
Forum List  »  NDB clusters

Re: performance
Posted by: Mikael Ronström
Date: October 06, 2017 11:52AM

Your table is partitioned on the column did.
So the query will scan all rows in the partition
to find the rows. The rows are not sorted in
any specific order with regards to the 'did' column.
So yes, this query will slow down with more rows in
the table since more rows have to be examined.

So a simple change that will speed up your query
substantially is to swap the order of the Primary Key
index from uniqueid,did to did,uniqueid. In this
case the ordered index of the primary key will
only check read rows with did=xyz.

Options: ReplyQuote

Written By
October 03, 2017 01:23AM
Re: performance
October 06, 2017 11:52AM

Sorry, you can't reply to this topic. It has been closed.

Content reproduced on this site is the property of the respective copyright holders. It is not reviewed in advance by Oracle and does not necessarily represent the opinion of Oracle or any other party.