Posted by Vadim |
…
Mmap/pread ratio
1
13.79
33.53
2.43
4
29.36
99.15
3.38
16
20.48
136.
89
6.68
64
20.34
137.98
6.78
256
20.69
137.97
6.67
So in conclusion memory …
Posted by Vadim |
… benchmark see here http://www.mysqlperformanceblog.
com/2006/06/13/quick-look-at-… ratio
1
335.56
429.
89
1.28
4
165.16
… ratio
1
344.72
429.
89
1.25
4
531.6
…
Posted by peter |
Looking at post by Konstantin Osipov we should finally get Query Cache working with prepared statements in MySQL 5.1
The interesting thing Konstantin notes it just took few days to fix it – I believe MySQL Support Team alone spent much more times …
Posted by peter |
…but what is interesting is the server line which shows nginx 0.5.32
I have checked and indeed images served from assets.hulu.
com have nginx server in response header.
This would be the first site of this (planned) scale to run Nginx. So far even…
Posted by peter |
…
| benchmark(100000000,1+2) |
+————————–+
| 0 |
+————————–+
1 row in set (18.
89 sec)
As you can see this is hell a lot of difference !
Depending on your application performance with …
Posted by Vadim |
…138 | 15127 | 0 | 0 | 51 | 17855 | index | PRIMARY | art | author87 |
|
89 | 1 | 3892 | 151 | 15088 | 0 | 0 | 51 | 17855 | index…
Posted by Vadim |
…
E.g. update_key benchmark from sysbench. There are also results with Yasufumi’s rw_locks (http://bugs.mysql.
com/bug.php?id=26442)
ThreadsStandard InnoDBGoogle smpfixYasufumi rw_lock
19700.2810601.969432.44
214355.6616673.3112783.58
416104.202669….
Posted by Baron Schwartz |
…20 0.35
site.clicks 1.26 3.05 5.13
site.actions_finished 1.14 0.46 0.74
site.ratings 0.91 0.
89 0.48
The difference is clear. The buffer pool contains over 8G of data for the site.benefits table on the active …
Posted by Vadim |
… 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88
89 90 91 92 93 94 95 96 97 98 99 100 done.
Some more information available on
http://www.percona.
com/docs/wiki/percona-xtradb:patch:innodb_expand_import
If you want some internals: I …
Posted by Vadim |
The feature I announced some time ago http://www.mysqlperformanceblog.
com/2009/06/08/impossible-possible-moving-innodb-tables-between-servers/ is now available … 75 76 77 78 79 80 81 82 83 84 85 86 87 88
89 90 91 92 93 94 95 96 97 98 99 100 done.
When …