I take rsync for granted. It's just the best way to keep stuff out there up to date with stuff over here, and vice versa. And lately, I've been using it to supplant my usage of scp. And it works. Brilliantly. And until recently, I hadn't stopped to realize: Hey, you know, this thing somehow works out differences between files out there and over here long before what's here gets there. I know, duh, but I just hadn't considered it.

Well, Paul Holbrook reminded me of this tonight, with links to Andrew Tridgell's paper on the algorithm, among other things. Damn, things like this remind me that I'm supposed to be getting my butt back into school...


Archived Comments

  • Also see the excellent The rsync Algorithm there is a mp3 of his presentation http://archive.linuxsymposium.org/ols2000/2000-07-21_15-02-49_C_64.mp3 Very nice. Wish they would get the newer symposiums up.
  • If I were you, I would take a serious look at Unison, which is like rsync, but it is two-way. I use it for all my backup/sync needs. http://www.cis.upenn.edu/~bcpierce/unison/