June 19, 2013

Post: Using MyISAM in production

MySQL could crash or go swapping. Second – table will be locked while repair is going. If this is frequently used tableuses table locks and has concurrent inserts which can go concurrently with selects. This is sometimes presented as great concurrency for inserts

Post: Fighting MySQL Replication Lag

…millions of simple updates instead of one large update to help MySQL replication lag it is not … using INSERTSELECT – which is in similar to what I just described but can be much worse as SELECT…function which will take SELECT query and the table to which store its result set and use in your …

Post: Identifying the load with the help of pt-query-digest and Percona Server

… can have a really large log file generated. … 1.00 0.00 INSERT poller_output # MISC 0xMISC …use to gather more data about the underlying tables involved and the query execution plan used by MySQL. The end resultselecting every column from all the tables involved in the query, probably selecting

Post: MySQL File System Fragmentation Benchmarks

tables and does specified number of inserts going to random tables. I used default MySQL settings for MyISAM (tabletables which could be because of extent allocation for rather large tables. For 10000 tables we had just 1000 of 4K rows in the table

Post: The perils of InnoDB with Debian and startup scripts

…buffer pool as large as we’d…result | insert … | 6351 | user | | NULL | 6355 | user | | NULL | 6356 | user | statistics | select … | 6357 | user | statistics | select … | 6358 | user | Sending data | select … | 6359 | user | statistics | select… 500 mysql tables in use 7, locked 0 MySQL thread…

Post: Using VIEW to reduce number of tables used

used on massive scale (blog hosting, forum hosting etc), resulting of hundreds of thousands if not millions of tablesused to ensure we do not insert data in the view which will be invisible. The VIEW approach works for SELECTtable MySQL

Post: Side load may massively impact your MySQL Performance

mysql-socket=/tmp/mysql_sandbox5516.sock prepare Running Sysbench and MySQLDump. Note we run them in the loop to see how resultmysql [localhost] {msandbox} (information_schema) > select…note MySQL actually uses midpoint insertion for…MySQL has buffer pool which can be easily washed away by large table

Post: Cache Performance Comparison

result from the cache and ship it back, so largeMySQL queries. On Selecting from the table I should note MySQL is rather fast selecting from the tableTABLE `test` ( `k` varchar(60) NOT NULL default ”, `val` varchar(255) default NULL, PRIMARY KEY (`k`) ) ENGINE=MyISAM DEFAULT CHARSET=utf8; INSERT

Post: Heikki Tuuri answers to Innodb questions, Part II

resultstable scan would be fast. Maybe the table has become fragmented? Or maybe you have not insertedlarge data manageable. MySQL Master Master Manager can help you to automate it a bit. I also like to keep tablesselect * from table where id=5 show innodb status: Hash table size 10624987, used

Post: Impact of the sort buffer size in MySQL

mysql test -e “insert into sorttest value (‘$UUID’);” let “COUNT=COUNT+1″ done I know, I could have used the uuid() function of MySQLuseful to interpret the results. The figure below shows a graphical representation of the results