Comments on: How fast can MySQL Process Data http://www.mysqlperformanceblog.com/2008/04/09/how-fast-can-mysql-process-data/ Everything about MySQL Performance Sat, 07 Nov 2009 18:35:44 -0800 http://wordpress.org/?v=2.8.4 hourly 1 By: ries http://www.mysqlperformanceblog.com/2008/04/09/how-fast-can-mysql-process-data/comment-page-1/#comment-273614 ries Fri, 11 Apr 2008 22:55:49 +0000 http://www.mysqlperformanceblog.com/2008/04/09/how-fast-can-mysql-process-data/#comment-273614 I see you use the ENGINE=MEMORY. This also means that the count(*) you are doing is not transactional save, this is what I understand... Which in return means that if you have multiple deletes/inserts/updates going on on your table you are never sure if the count(...) value you get is correct at that point in time?? Also I do feel doing doing a count(*) is a weird form of 'testing' I think what you do a bit below doing the aggregates is more valid because then MySQL really needs to scan the table rather then looking at some index (if that is optimized). I am also wondering why you didn't use an innodb storage engine and using the MEMORY storage type. I think every serious database user want's to have something that is transactional save, right?? just my two cents I see you use the ENGINE=MEMORY. This also means that the count(*) you are doing is not transactional save, this is what I understand…
Which in return means that if you have multiple deletes/inserts/updates going on on your table you are never sure if the count(…) value you get is correct at that point in time??
Also I do feel doing doing a count(*) is a weird form of ‘testing’ I think what you do a bit below doing the aggregates is more valid because then MySQL really needs to scan the table rather then looking at some index (if that is optimized).

I am also wondering why you didn’t use an innodb storage engine and using the MEMORY storage type. I think every serious database user want’s to have something that is transactional save, right??

just my two cents

]]>