When backing up a large filesystem over a network for the first time ,
I always use tar | rsh tar to populate the backup because I have found
it to be much faster than rsync.
I have always wondered why rsync is so much slower than tar in this
regard, perhaps this could be investigated. I have never tried to tune
rsync for a fresh load, but perhaps making rsync behave just like tar
for the first push of data might be a good idea.
Wayne Davison wrote:
On Sun, Sep 11, 2005 at 06:21:13PM -0400, Matt McCutchen wrote:
Since both tar and rsync read and write filesystems in great detail,
they have many analogous sections of source code.
An interesting idea. Rsync 2.6.6 now has the --only-write-batch option
which can allow it to function like "tar c" when a copy is made into an
empty directory. That means that it can be thought of as a superset of
tar's basic functionality (though it does not have all of tar's options
nor is it compatible with tar).
Someone may want to look into creating some library functions that tar
and rsync could share. But having one program do both jobs may be
problematical if backward-compatibility with older tar versions was
desired. However, someone designing a next-generation file-transfer
tool my well want to make it easy to also do tar's job, and more
advanced archive operations, as you suggested.
..wayne..
|
--
To unsubscribe or change options: https://lists.samba.org/mailman/listinfo/rsync
Before posting, read: http://www.catb.org/~esr/faqs/smart-questions.html
