http://my.oschina.net/davehe/blog/174662
RDB-
In binary files that exist in Dump.rdb
Dump the whole DB, when the data are many, not suitable for frequent storage, save the time interval should be longer
Save data Files Compact, recover Big data files fast
If save, shutdown, slave before crash, then the middle of the operation is not able to recover.
The only thing the parent process has to do when saving the RDB file is to fork out a child process, and the child process will handle all subsequent save work, and the parent process does not need to perform any disk I/O.
RDB Snapshot -automatically saves a dataset when "there are at least 1000 keys changed in 60 Seconds": Save 60 1000 "
AOF-
The write instructions are appended to the end of the file and are continuously written to a log file like
The default policy for AOF is fsync once per second
APpendfsync always executes every time .
appendfsync everysec Default synchronous operation per second (recommended, default)Appendfsync No does not actively synchronize, by the operating system to do, 30 seconds once
[Redis] RDB & AOF