Feature request: block de-duplication #46
Description
I am a completly noob when it comes to FUSE and have only a basic academic
understanding of file systems, but I would like to contribute to s3backer
making some improvement as my final graduation project. I was thinking about
implementing some basic deduplication mechanism, where the dedup indexes are
stored on the client machine, like the cache file. If think this could improve
s3 storage usage and make s3backer even more cost effective.
I was also thinking about using s3backer as an archiving solution for a
production environment. As I would be storing only cold data, eventual
consistency would not be problem, as the time between a possible read and its
predecessor write would be too large. Is there any scenario where I can get
data corrupted using the data cache inside a persistent storage device? Let´s
not consider the case where this very own device fails.
Thanks in advance,
Original issue reported on code.google.com by vovoh...@gmail.com on 25 Mar 2014 at 1:01