June 19, 2013

Post: Galera Flow Control in Percona XtraDB Cluster for MySQL

… a six-hour tutorial about Percona XtraDB Cluster (PXC) for MySQL.  I actually had more material than I covered (by design… replication feedback mechanism, unlike anything you find in standard async MySQL replication. It is my belief that the lack of understanding… lookups into the certification index, which is a hash table.  A small fc_limit does however keep the certification index smaller in memory…

Post: Announcing Percona XtraBackup 2.1.1 GA

… rebuilding secondary indexes on innobackupex –apply-log –rebuild-indexes. This allows parallel processing of individual tables when rebuilding the index. Percona XtraBackup… now uses Perl’s DBD::MySQL package for server communication instead of spawning the MySQL command line client. Support for…

Post: MySQL and Percona Server in LinkBench benchmark

… related to BUG#67808. For MySQL 5.6.11 top mutexes is &buf_pool->mutex,&new_index->lock. I profiled 5.6…_capacity = 5000 sync_binlog=0 max_connections=5000 table_open_cache=5000 table-definition-cache=1000 query_cache_size=0 query_cache…_instances=16 # MySQL 5.5 and 5.6 loose-innodb_io_capacity_max = 15000 #Percona only innodb_adaptive_hash_index_partitions=8…

Post: Percona Server 5.5.30-30.2 rerelease fixes non-restart issue

… release of Percona Server for MySQL has fixed a number of performance bugs. (Alexey Kopytov) Drop table performance feature has been removed… explicitly named foreign key constraint which implied an implicit secondary index with the same name. Fixed by detecting such cases and… (Alexey Bychko). Percona Server for MySQL was missing help texts in the MySQL client because the help tables were missing. Bug fixed #1041981…

Post: Is Synchronous Replication right for your app?

… single row (well, the PRIMARY KEY index entry for that row).  This means … MySQL replication from this instance, since MySQL replication is asynchronous. What about semi-sync MySQL replication… maintained in separate tables and there also exists a users_groups table to define the relationship …

Comment: MySQL Partitioning - can save you or kill you

… single-row writes are likely to be slower in PARTITIONed tables unless you include something to facilitate pruning. A billion rows…. * A ‘range’ delete would mostly follow above two points. * Secondary indexes are also BTrees, so they need updating (although delayed). Yes… code) here: http://mysql.rjweb.org/doc.php/partitionmaint The third use case for PARTITIONing involves 2D-like indexing. I have not…

Post: Benchmarking Percona Server TokuDB vs InnoDB

…) DEFAULT ” NOT NULL, PRIMARY KEY (id) ) CREATE INDEX k on sbtest$I(k) and tables sbtest$I_r10, sbtest$I_r100, sbtest…: sysbench –test=insert_roll.lua –oltp-table-size=10000 –mysql-user=root –oltp-tables-count=32 –mysql_table_engine=tokudb –oltp_auto_inc=on –max…_threads = 1 myisam_recover socket=/var/lib/mysql/mysql.sock user=root skip-grant-tables TokuDB-related options are all defaults, as I…

Comment: MySQL Partitioning - can save you or kill you

… quicker on a partitioned table, as it will be adding to a smaller index, and a smaller ‘chunk’ od table. Depending on how often you write to the table, you may not… other column, these could easily be quicker on a partitioned table, as Mysql can fire off one thread per partition (at least…

Comment: How to recover table structure from InnoDB dictionary

… then ibdata1 should be valid, maybe with some corruption. The tables structure you can get from the dictionary as it is…/innodb-data-recovery-tool:mysql-data-recovery:start . The biggest challenge here is to find index_id for each table – http://www.percona.com/docs/wiki/innodb-data-recovery-tool:mysql-data-recovery:advanced…

Post: MySQL Indexing Best Practices: Webinar Questions Followup

I had a lot of questions on my MySQL Indexing: Best Practices Webinar (both recording and slides are available now) … details. Q: how mysql use index for group by? A: If you have Index on the column MySQL can avoid temporary table or filesort for group by by this column. This works because by scanning data in index order MySQL