[Box Backup-dev] Re: [Box Backup] Re: Block Sizes and Diffing

Chris Wilson boxbackup-dev at fluffy.co.uk
Tue Oct 2 19:39:21 BST 2007


Hi Ben,

On Wed, 26 Sep 2007, Ben Summers wrote:

> Consider a 128k text file which uses 4k blocks. Insert a character 
> somewhere in the middle of a block, then back it up again. You've now 
> got a gap which isn't covered by any of the existing blocks, and it's 
> 4097 bytes long. How can a non-standard block size be avoided without 
> bandwidth inefficiency or knowledge of the key on the server side? What 
> do you do when it gets lots of little inserts over hundreds of backup 
> runs?
>
> Benefits have costs. Tradeoffs have to be made.

OK, thanks for the explanation. I didn't realise the block scanning 
algorithm was clever enough to deal with this case :-)

How about reducing the number of re-reads by computing checksums of all 
blocks at all sizes with a single pass?

Cheers, Chris.
-- 
_____ __     _
\  __/ / ,__(_)_  | Chris Wilson <0000 at qwirx.com> - Cambs UK |
/ (_/ ,\/ _/ /_ \ | Security/C/C++/Java/Perl/SQL/HTML Developer |
\ _/_/_/_//_/___/ | We are GNU-free your mind-and your software |



More information about the Boxbackup-dev mailing list