Editor
6
edits
Difference between revisions of "Performance tuning"
Jump to navigation
Jump to search
Copy a warning from the source about skip-innodb_doublewrite, which may not be safe. →Database workloads
m (→Whole Disks versus Partitions: more clarification on illumos write cache enable) |
(Copy a warning from the source about skip-innodb_doublewrite, which may not be safe. →Database workloads) |
||
| Line 167: | Line 167: | ||
Make separate datasets for InnoDB's data files and log files. Set recordsize=16K on InnoDB's data files to avoid expensive partial record writes and leave recordsize=128K on the log files. Set primarycache=metadata on both to prefer InnoDB's caching.[https://www.patpro.net/blog/index.php/2014/03/09/2617-mysql-on-zfs-on-freebsd/] Set logbias=throughput on the data to stop ZIL from writing twice. | Make separate datasets for InnoDB's data files and log files. Set recordsize=16K on InnoDB's data files to avoid expensive partial record writes and leave recordsize=128K on the log files. Set primarycache=metadata on both to prefer InnoDB's caching.[https://www.patpro.net/blog/index.php/2014/03/09/2617-mysql-on-zfs-on-freebsd/] Set logbias=throughput on the data to stop ZIL from writing twice. | ||
Set skip-innodb_doublewrite in my.cnf to prevent innodb from also writing twice. The double writes are a data integrity feature meant to protect against partially-written pages, but those are not possible on ZFS | Set skip-innodb_doublewrite in my.cnf to prevent innodb from also writing twice. The double writes are a data integrity feature meant to protect against partially-written pages, but those are not possible on ZFS.[https://www.percona.com/blog/2014/05/23/improve-innodb-performance-write-bound-loads/] WARNING: The source now says, "Update: do not do this, this has been proven to corrupt data!" | ||
==== MyISAM ==== | ==== MyISAM ==== | ||