… = 1600M innodb_log_buffer_size = 16M innodb_flush_log_at_trx_commit = 1 innodb_lock_wait_timeout = 300 innodb_thread_concurrency = 0 innodb_sync_spin_loops = 40 innodb_locks_unsafe_for_binlog = 1 innodb_max_dirty_pages_pct=15 innodb_support…
Comment: kernel_mutex problem. Or double throughput with single variable
December 4, 2011 By 17 Comments
Setting innodb_sync_spin_loops is very interesting discussion because there is really no “right” … the limiting mutex for your workload the different amount of spinning might make sense. Better solution would be to have this… makes sense to wait. For example if we run long spin and can discover we either get the mutex after 10us…
Comment: MySQL on Amazon RDS part 2: Determining Peak Throughput
March 28, 2011 By 4 Comments
… OFF innodb_spin_wait_delay 6 innodb_stats_on_metadata ON innodb_stats_sample_pages 8 innodb_strict_mode OFF innodb_support_xa ON innodb_sync_spin_loops 30 innodb_table_locks ON innodb_thread_concurrency 0 innodb_thread…
Comment: SHOW INNODB STATUS walk through
June 27, 2007 By 40 Comments
… waits 29544296 RW-shared spins 17771305, OS waits 8349871; RW-excl spins 10508828, OS waits 5276204 Currently I have innodb_sync_spin_loops set to the…
Comment: Returning to InnoDB scalability
August 31, 2006 By 42 Comments
Would reducing innodb_sync_spin_loops below its default of 20 be helpful here? With the patch there are still 2500 spin rounds per OS wait, so it looks as though it’s still burning a lot of CPU.
Comment: kernel_mutex problem. Or double throughput with single variable
December 16, 2011 By 17 Comments
… ago, that I could not get to perform well in innodb. It seemed like MySQL would attack one thread, and starve… out the old code and data, and ran it with innodb_sync_spin_loops=64, and the workload performed much better. Thanks again, and…
Comment: Returning to InnoDB scalability
August 31, 2006 By 42 Comments
James, I don’t think innodb_sync_spin_loops will help a lot, but you might try 
Btw, how are results with patch and without ?
Comment: kernel_mutex problem cont. Or triple your throughput
December 19, 2011 By 10 Comments
We recently tried some of this tuning to get rid of some contention that we are having. sync_spin_loop changes made no difference, and decreasing innodb_thread_concurrency to 16 or under actually caused our site to crash. So obviously this stuff is work-load dependent.
Post: Understand InnoDB spin waits, win a Percona Live ticket
September 2, 2011 By 2 Comments
…INNODB STATUS mean? Mutex spin waits 5870888, rounds 19812448, OS waits 375285 To understand this text, you have to understand how InnoDB…The related source files are in the sync/ InnoDB source code directory if you want … a spin-wait. rounds 19812448 is the number of times threads looped in the spin-…
Comment: read_ahead (disabled) as steroid
May 12, 2009 By 4 Comments
… in spin loop should increase performance of mutex/rw_lock contention. It may affect 5~10% in throughput. – log_buffer_flush_maybe_sync() will decrease the number of log sync at innodb_flush_log_at_trx_commit=[0|2…. – relax condition to flush log and contract ibuf at inner loop of srv_master_thread() – if (n_pend_ios < 3 && (n…
