Jeff, >> Has anyone any benchmark between 4.0 and 4.1? >> >> MySQL website says that version 4.0 has some performance >> improvements but I´ve heard that MySQL-4.0 is 20% faster >> than MySQL-4.1. > > As far as what? Thats a very general statement. How do you even know that > affects you?
The following thread appeared in the FreeBSD list: --------------------------- > Here are the results of the brief benchmarks that I ran. I posted the > best and worse results out of running each test 5 times. > > ----------------- > System: > > Dual Opteron 244s > RAM 1GB > 3 x 36GB U320 SCSI RAID-5 on Adaptec 2120s > > ----------------- > OS: > > FreeBSD <snip> 5.3-RELEASE FreeBSD 5.3-RELEASE #2: Mon Nov 15 10:35:13 > PST 2004 <snip> i386 > > ----------------------------------------------------------- > MySQL 4.1.7 WITH_OPENSSL BUILD_OPTIMIZED > ----------------------------------------------------------- > master# super-smack update-select.smack 30 10000 > Query Barrel Report for client smacker > connect: max=15ms min=4ms avg= 8ms from 30 clients > Query_type num_queries max_time min_time q_per_s > select_index 300000 3 0 4365.04 > update_index 300000 4 0 4365.04 > > master# super-smack update-select.smack 30 10000 > Query Barrel Report for client smacker > connect: max=12ms min=0ms avg= 7ms from 30 clients > Query_type num_queries max_time min_time q_per_s > select_index 300000 3 0 4799.58 > update_index 300000 3 0 4799.58 > > master# super-smack select-key.smack 30 10000 > Query Barrel Report for client smacker1 > connect: max=10ms min=7ms avg= 9ms from 30 clients > Query_type num_queries max_time min_time q_per_s > select_index 600000 3 0 10614.75 > > master# super-smack select-key.smack 30 10000 > Query Barrel Report for client smacker1 > connect: max=10ms min=6ms avg= 7ms from 30 clients > Query_type num_queries max_time min_time q_per_s > select_index 600000 4 0 10666.14 > > ----------------------------------------------------------- > MySQL 4.0.22 WITH_OPENSSL BUILD_OPTIMIZED > ----------------------------------------------------------- > master# super-smack update-select.smack 30 10000 > Query Barrel Report for client smacker > connect: max=13ms min=9ms avg= 11ms from 30 clients > Query_type num_queries max_time min_time q_per_s > select_index 300000 3 0 4839.98 > update_index 300000 3 0 4839.98 > > master# super-smack update-select.smack 30 10000 > Query Barrel Report for client smacker > connect: max=22ms min=0ms avg= 17ms from 30 clients > Query_type num_queries max_time min_time q_per_s > select_index 300000 3 0 4963.34 > update_index 300000 2 0 4963.34 > > master# super-smack select-key.smack 30 10000 > Query Barrel Report for client smacker1 > connect: max=19ms min=5ms avg= 9ms from 30 clients > Query_type num_queries max_time min_time q_per_s > select_index 600000 2 0 12387.57 > > master# super-smack select-key.smack 30 10000 > Query Barrel Report for client smacker1 > connect: max=15ms min=0ms avg= 7ms from 30 clients > Query_type num_queries max_time min_time q_per_s > select_index 600000 2 0 13201.30 > > -------------------------------------------------------------- > > It seems 4.0.22 is faster in each benchmark. Around a 20% increase. I > wonder why this is? --------------------------- Once I plan to run FreeBSD-5.3 + MySQL-4.1.7 in my server, it can affect me. Ronan -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]