June 19, 2013

Post: Choosing innodb_buffer_pool_size

… MySQL needs – these include MySQL buffers – query cache, key_buffer, mysql threads, temporary tables, per thread sort buffer which can be allocated. There… 1GB good estimates for this number. Eliminate Double Buffering – This is again very important for buffer pool size choice. You do not want… spike. Plus not all kernels work quite well with swap disabled and there are other reasons against it. For some people…

Post: Innodb Double Write

… tablespace it is recovered from double write buffer. How much does double write buffer affect MySQL Performance ? Even though double write requires each page written… guarantees you no partial page writes could exist you can disable doublewrite by setting innodb_doublewrite=0 It is however not…

Post: Heikki Tuuri Innodb answers - Part I

… optimization. I frequently see batch jobs killing server performance overtaking buffer pool. Though full table scan is only one of replacement… tests by Peter Zaitsev, O_DIRECT works well to remove double buffering. In distant future, I would like to have the memlock… for semaphores in SHOW INNODB STATUS), you can try to disable thread throttling completely by setting the value 0. But if…

Post: Is there room for more MySQL IO Optimization?

… most cases – it makes sure there is no overhead of double buffering and I can save the limited amount of file system… MySQL 5.6) so I tried that as well as disabling read ahead all together. In both cases for this given… buffered and unbuffered IO for read only workloads, if anything unbuffered performance should be better as it has less overhead with double buffering

Post: MySQL Performance on Memory Appliance

…/sec with multiple threads which corresponds to even worse scalability. Disabling adaptive hash index could take us to 14.500 req… percents. I also tried running with key buffer on and off. Interesting enough disabling key buffer actually worked better for this workload avoiding… conditions I got about 13.5K queries/sec from MyISAM (double of Innodb already) which scaled to 70.000 queries/sec…

Post: xtrabackup-0.7 (RC)

… files – use posix fadvise call to disable OS caching of copying files – disable recovery of double buffer Also we added binary builds for FreeBSD…

Post: Updated msl (microslow) patch, installation walk-through!

… VISUAL=/bin/cat mysqlbug | grep ‘Configure command’ Configure command: ./configure ‘–disable-shared’ ‘–with-server-suffix=-community’ ‘–without-embedded-server’ ‘–with-ndbcluster… -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector –param=ssp-buffer-size=4 -m64 -mtune=generic’ ‘CXX=gcc’ ‘LDFLAGS=’ This command… should put double quotes around the entire string of comma-separated flags. Currently “microtime” is mandatory meaning you cannot disable it…

Post: MySQL 5.5.8 - in search of stability

… using this formula: (100*Innodb_buffer_pool_pages_dirty)/(1+Innodb_buffer_pool_pages_data+Innodb_buffer_pool_pages_free). This is…_flush_method = O_DIRECT innodb_log_buffer_size = 64M innodb_buffer_pool_size = 26G innodb_buffer_pool_instances=16 innodb_log_file_size… doubles the amount of writes :) , as you see from its name. So, let’s see what result we have when we disable

Comment: Jeremy Cole on MySQL Replication

… is aproximately correct. The data is modified directly in the buffer pool pages, which becomes “dirty” if there are some unflushed… via one more “physical” log file which is called “double write buffer” to handle problem of partial page writes. The log file…/19/group-commit-and-xa/ – disabling XA saves you from extra fsync which allows to double results but results still do not…

Post: Review of MySQL 5.6 Defaults Changes

… recommend adding support for value 0 which would mean limit disabled. Very few users actually need DOS prevention this variable strives… simple and complicated queries efficiently automatically. join_buffer_size = 256k The default is now double of what it was previously. I wonder… or is it just looking to reduce number of different buffer sizes used. sync_master_info = 10000 Quite useless change in…