it doesn't because it doesn't address that issue at all. this is simply an issue of how does one store the data reliably when its barely all you can do to store it once.
now that erasure codes have freed us from the tyranny of RAID, folks seem empowered to store prodigious amounts of data. BUT, this Big Data comes with different expectations, one of which is that there are no backups (in a user visible sense). almost no one can afford the traditional form of backups (as completely separate copies). On Sep 14, 2013, at 10:10 PM, David Lang wrote: > On Sat, 14 Sep 2013, Adam Levin wrote: > >> The other system we saw recently is called Amplidata. They have a file >> system (if it can really be called that) that provides so much protection >> via erasure coding that you no longer need to back up the data because the >> calculated protection is three orders of magnitude better than what >> replicated backups can provide. > > so how does this protect against 'rm -rf /' or something that overwrites the > data (or corrupts the filesystem) > > David Lang_______________________________________________ > Tech mailing list > Tech@lists.lopsa.org > https://lists.lopsa.org/cgi-bin/mailman/listinfo/tech > This list provided by the League of Professional System Administrators > http://lopsa.org/ > _______________________________________________ > tachyon mailing list > tach...@research.att.com > http://mailman.research.att.com/mailman/listinfo/tachyon ----------------------- Andrew Hume 949-707-1964 (VO and best) 732-420-2275 (NJ) and...@research.att.com
_______________________________________________ Tech mailing list Tech@lists.lopsa.org https://lists.lopsa.org/cgi-bin/mailman/listinfo/tech This list provided by the League of Professional System Administrators http://lopsa.org/
