Nesh, > With all of the benchmark test I have used the same DB structure, same > TABLE structure, and finally same iterations. > > This probably means that Solaris I/O is really poor, or there is some other > explanation (like changing some kernel parameters to get Solaris working > well with DB kind of stuff). > > At the moment I am trying to find out exactly what is the problem. It > should not be MySQL itself because the first 2 boxes have different results > from the last 2 boxes. This leads me to the conclusion that Solaris is doing > something weird with the I/O. Unfortunately, I do not have Solaris 8 to test > MySQL in that environment. > > Note that all the boxes are running the same MySQL version so that could > not be the problem (3.23.47).
Interesting statistics! What else were the m/user, m/tasking OpSys doing at the time? How much 'tuning' of the configs/dbs had taken place in each case? Can you compare the results without the 'I/O', by using native-MySQL on each server, locally? Regards, =dn --------------------------------------------------------------------- Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php