…. An additional optimization we have for SSDs is big log files. For this run, I used innodb_log_file_size=4G (innodb_log_files_in_group=2) for Percona…:autoextend innodb_file_per_table=1 innodb_flush_log_at_trx_commit=2 innodb_log_buffer_size=8M innodb_log_files_in_group=2 innodb_log_file_size=4G innodb_log_block_size=4096 innodb…
Post: InnoDB, InnoDB-plugin vs XtraDB on fast storage
January 13, 2010 By 13 Comments
…configured in software RAID0 to store InnoDB datafiles. For InnoDB logs and system tablespace I used …yoshinorimatsunobu.blogspot.com/2009/05/tables-on-ssd-redobinlogsystem.html and taking fact that …about 88% worse than all files on FusionIO) To summarize MySQL InnoDB/InnoDB-plugin/XtraDB is not …
Comment: MySQL 5.6.7-RC in tpcc-mysql benchmark
October 9, 2012 By 22 Comments
… results as well. Expectations for innodb_log_file_size should be changing. Simon Mudd using 32GB of redo log file on a 192G box are the way things are going, particularly for people who want to maximise SSD life… be fine with 4G or less. The optimal value for innodb_buffer_pool_instances depends in part on the distribution of…
Comment: Percona Server vs MySQL on Intel 320 SSD
February 23, 2012 By 21 Comments
… usually a symptom of innodb_io_capacity set too low for the workload and system. For an SSD setup with high write… server dirty page flushing respects the innodb_io_capacity setting, Percona server flushing doesn’t. Set innodb_io_capacity too low and… 85% of the InnoDB log file space used, while Percona server won’t. You’ll also normally benefit from setting innodb_purge_threads to…
Comment: Recovery beyond data restore
August 3, 2008 By 11 Comments
… reply Peter. I did a lot of research on Flash SSD and came to the conclusion that random writes are often worse on SSDs than on conventional hard drives. I’m also not seeing… transactions aren’t written to a physical disk and the InnoDB log files are apparently the bottleneck here based on my testing today…
Comment: MySQL Backup tools used by Percona Remote DBA for MySQL
March 17, 2013 By 8 Comments
…. Some factors that affect speed of restore include innodb_log_file_size, innodb_buffer_pool_size, innodb insert buffer (default to 50% of buffer pool… are also several other innodb setting that affect the number of background threads doing IO etc. With SSD‘s in particular using…
Comment: Benchmarks of new innodb_flush_neighbor_pages
July 3, 2012 By 8 Comments
… than server configuration” [End Quote] This is another article on log file sizes and stalls/lockups, even though the lock issue might…/05/24/binary-log-file-size-matters/ [Quote] “Here’s what we did: we have reduced the size of binary log file from default… to raid 10 a bunch of SSD‘s then we could put the Innodb straight to SSD and not even use any Ram…
Post: Effect of adaptive_flushing
December 4, 2009 By 5 Comments
… more writes per second == shorter lifetime, and so a shorter SSD lifetime is a negative side effect of enabling adaptive flushing… to control InnoDB‘s checkpoint age is by adjusting the log file size. This server is configured with 2 * 1.5G log files, and innodb_io_capacity=1000. While adaptive flushing was off, the checkpoint age was comparable to the log file size, however…
Comment: Heikki Tuuri to answer your in depth Innodb questions
September 29, 2007 By 51 Comments
… using solid state disks for storing ib_log* files I know that innodb writes files in the log file group in a round robin fashion. I… the first file in a log file group. My questions are: 1) What is always being written to the first file in the log file group… first file in the log file group and the SSD dynamic wear leveling algorithm. I’m also wondering if you would recommend using SSD for…
Post: Optimizing InnoDB for creating 30,000 tables (and nothing else)
December 22, 2011 By 5 Comments
…: innodb-flush-method, sync-frm and innodb-flush-log-at-trx-commit. There is an undocumented option for innodb-flush…() call when extending the data file, so I tried setting a higher innodb-autoextend-increment and a larger… a “I’m glad this isn’t a SSD” killer amount of fsync() activity, it took a…
