Mydumper is a high-performance multi-threaded backup and Recovery tool for MySQL and drizzle.
Main Features:
1, lightweight, written in C
2, the execution speed is mysqldump 10 times times
3. Consistent snapshots of transactional and non-transactional tables (more than 0.22 versions)
4, file compression (in the form of. GZ)
5, Support export Binlog (this I tested, but did not succeed, advice, test version 0.23)
6, multithreading recovery (more than 0.21 version)
7, the daemon process mode of operation, timed snapshots and continuous binary log (0.50 version above)
8, the most pleasing is: Open source (GNU GPLv3)
Installation section:
Please refer to official documentation: Http://docs.mydumper.org/compiling.html (this is 0.2 document)
After the installation is successful, there will be a mydumper (backup) myload (Restore) two tools
Parameters can be referred to: Mydumper--help detailed Explanation: http://docs.mydumper.org/mydumper_usage.html
Example: Mydumper-b monster-o/tmp/abc/test1-c-V 3
More Wonderful content: http://www.bianceng.cnhttp://www.bianceng.cn/soft/tools/
Three types of files are generated: *schema.sql structure file *.sql data file. Metadata metadata file (this is the binary log log location that contains the backup time, such as: Started Dump at:2012-07-27 19:00:38
Show MASTER STATUS:
log:mysql-bin.000019
pos:542237180)
The basic use of Myloader is:
Myloader--directory/path/to/mydumper/backup "Options"
Mydumper Partial parameter explanation:
Personal view of the more practical parameters:
--threads number of backup data threads, defaults to 4 (PS: not more threads the better, the actual test time commands)
--long-query-guard set the upper limit for long queries by default is 60S,
--kill-long-queries if the long query exceeds the specified time, the default mydumper will exit, using this parameter, you can directly kill the long query, to ensure that the backup continues.
--regex matches multiple databases or datasheets
--rows a few tables into several chunk, you can divide a table into several files, the number of files is based on the number of rows and rows of the table to allocate.
(PS the tool does not support character set settings; there is no option like Myslqdump--master-info to write the change master to statement to the SQL file; The tool is consistent and doing well!) )