Selon Behrang Saeedzadeh <[EMAIL PROTECTED]>: > Mathias, > > Thanks a lot! > > > I will not explain the same thing for sqlserver, sybase ..., but when > > your RDBMs > > have a data dictionnary, you don't need to execute count(*) :o) WITH > > Updated > > statistics of course. > > I'm a little bit confused here. Why the count(*) is not transformed to a > select from the data dictionary if this way is faster? And what's the > difference between updated statistics and statistics not updated? > > > With information_schema in 5.x and higher, innodb will act as it's done > > in all > > the other RDBMS. > > > > Hope that helps > > Sure! It helped by orders of magnitured more than I thought it can help ;-) > > > :o) > > Mathias > > > > -- > Behrang Saeedzadeh > http://www.jroller.com/page/behrangsa > > Using Opera's revolutionary e-mail client >
Well, The information in data dictionnary are correct only just after updating them. imagine at 12h, you update statistics, num_rows=2000. At 12h05, you insert 1000 lignes and delete 500. At 12h10, you ask the data dictinary num_rows, it will give you 2000, even if they are 2500. Hope that helps :o) Mathias -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
