June 18, 2013

Post: Benchmarking Percona Server TokuDB vs InnoDB

…NULL DEFAULT ’0′, PRIMARY KEY (`hid`,`mid`,`id`) ) and transactions are: $HID=rand(0,10000) // distributed by uniform distribution… innodb_flush_method = O_DIRECT innodb_log_buffer_size = 256M innodb_flush_log_at_trx_commit = 1 innodb_buffer_pool_size = 36G innodb_log_file_size = 4G innodb_log

Post: More on MySQL transaction descriptors optimization

…are, by definition, read-only transactions, so the server deals with an empty transaction list when creating read views. However… innodb_flush_log_at_trx_commit = 2 innodb_flush_method = O_DIRECT innodb_log_buffer_size = 16M innodb_buffer_pool_size = 52G innodb_log_file_size = 2000M innodb_log

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

… instead of failing or returning immediately if there is a transaction that executed a query which opened that table. Bug fixed… the fix for unnecessary log_flush_order_mutex acquisition. Bug fixed #1163262 (Alexey Kopytov). When mysqldump was used with –innodb-optimize-keys and…

Post: Percona Server for MySQL 5.5.30-30.2 now available

… instead of failing or returning immediately if there is a transaction that executed a query which opened that table. Bug fixed… the fix for unnecessary log_flush_order_mutex acquisition. Bug fixed #1163262 (Alexey Kopytov). When mysqldump was used with –innodb-optimize-keys and…

Post: Percona XtraBackup 2.0.7 for MySQL available for download

… that are caused by the log records in the transactional log being overwritten before they are copied by the log copying thread. Percona XtraBackup… all InnoDB error or diagnostic messages are never printed by xtrabackup_56. Bug fixed #1169971. innobackupex would still run with FLUSH TABLES…

Post: Percona XtraBackup 2.1.0 'release candidate' for MySQL available for download

… that are caused by the log records in the transactional log being overwritten before they are copied by the log copying thread. Percona XtraBackup… all InnoDB error or diagnostic messages are never printed by xtrabackup_56. Bug fixed #1169971. innobackupex would still run with FLUSH TABLES…

Post: Is Synchronous Replication right for your app?

… the transaction commits and that takes an fsync to the redo log by default, so applying Callaghan’s law to single-server Innodb… You can obviously relax that by simply not fsyncing every transaction (innodb_flush_log_at_trx_commit != 1), or work around it with by… your hotspots were really bad in standalone Innodb, you might consider relaxing the fsync:  set innodb_flush_log_at_trx_commit to something besides…

Post: Virident vCache vs. FlashCache: Part 2

… We see the exact same pattern when looking at transaction latency; the baseline numbers are roughly identical for …innodb_flush_method = O_DIRECT innodb_log_buffer_size = 128M innodb_flush_log_at_trx_commit = 1 innodb_log_file_size = 1G innodb_log_files_in_group = 2 innodb_purge_threads = 1 innodb

Post: How InnoDB handles REDO logging

…that affects both response time and throughput – The transaction log. The InnoDB transaction log handles REDO logging, this is the mechanism that provides the A (… at which InnoDB can log. You can request that InnoDB force its REDO records to the transaction log but not actually flush the log to …

Post: InnoDB Flushing: Theory and solutions

… survive a crash or system failure, InnoDB is logging changes into InnoDB transaction logs. The size of the InnoDB transaction log defines how many changed blocks we can… age of the oldest modified non-flushed page. Checkpoint_age is located somewhere between 0 and innodb_log_file_size. Point 0 means…