Announcing TokuDB v7.5: Read Free Replication

Today we released TokuDB® v7.5, the latest version of Tokutek’s storage engine for MySQL and MariaDB.

I’ll be publishing two blogs next week to go into more details about our new “Read Free Replication”, but here are high level descriptions of the most important new features.

Read Free Replication
TokuDB replication slaves can now be configured to process the binary logs with virtually no read IO. This is accomplished via two new server parameters: one to allow the skipping of uniqueness checks (for inserts and updates), the other to eliminate read-before-write behavior (for updates and deletes). The two other conditions are that the slave must be in read-only mode and replication must be row based.
Hot Backup Now Supports Multiple Directories (Enterprise Edition)
The original implementation of our Hot Backup functionality was only capable of backing up servers with single “data” directory for all files (datadir). This enhancement adds the capability to backup differing tokudb_data_dir, tokudb_log_dir, and binary log folder.
Increased Usage of Bulk Fetch
Long ago, we added the capability in TokuDB to utilize a bulk fetch algorithm when selecting ranges of rows, which significantly boosts the performance of certain SELECT statements. This release bring the same performance enhancement to more complex SELECT statements as well as DELETE statements. The new SELECT statements covered are “INSERT [IGNORE] INTO … SELECT”, “INSERT INTO … ON DUPLICATE KEY UPDATE”, “REPLACE INTO … SELECT”, and “CREATE TABLE … SELECT”.
Updated MySQL and MariaDB
We are committed to updating our MySQL and MariaDB versions with each release, so we are shipping TokuDB v7.5 on MySQL 5.5.39 and MariaDB 5.5.39.
To learn more about TokuDB:
Download it, check out the documentation, or see the full release notes.

Share this post

Comment (1)

  • Mark Callaghan Reply

    I am impressed. Read-free replication is a big deal. Now a disk-based slave (less expensive) can keep up with a pure-flash master or scale-out slaves will save more IOPs to be used for scale-out reads.

    September 19, 2014 at 11:27 am

Leave a Reply