On Mar 31, 2005, at 4:32 PM, Thomas Denier wrote:
We recently restored a large mail server. We restored about nine million files with a total size of about ninety gigabytes. These were read from nine 3490 K tapes. The node we were restoring is the only node using the storage pool involved. We ran three parallel streams. The restore took just over 24 hours.
The client is Intel Linux with 5.2.3.0 client code. The server is mainframe Linux with 5.2.2.0 server code. ...
I noticed that you didn't mention the file system type. The effects of file system type and layout of the subject instance is an often overlooked contributor to performance in operations which are mass-populating the file system, as a restoral will. A journaled file system can exhibit a lot of overhead as its journal is written with at least metadata, depending upon type; and an ill-located journal can make for a lot of disk arm diversions during the restoral, aggravating elapsed time.
IBM's outstanding documentation store includes a great series on Linux file systems, which one can jump into at http://www-106.ibm.com/developerworks/linux/library/l-fs7.html .
Richard Sims
