[Cryptography] Cubbit

[Tony Patti]

> Jerry Leichter <leichter at lrw.com> writes:
> 
> > The description Cubbit provides says that they distribute 36 copies in 
> > a Reed-Solomon code that allows recovery from any 24 copies.  But one 
> > way or another ... 36 copies requires that, somewhere, there be 35 
> > times the space of the original copy to provide the redundancy.
> 
>Nemo <nemo at self-evident.org> writes:
> 
> We have a RAID6 at work that uses 12 1TB drives. It provides a full 10TB
of usable storage, and any two of those 12 drives can fail without causing
us any data loss whatsoever.
> 
> How does this fit with your analysis?
>
>  - Nemo

RAID5 and especially RAID6 have TERRIBLE write performance, which you can
quantify here: https://wintelguy.com/raidperf.pl 
Disks are already the slowest part of a system, so why make their
performance even less?
I always use RAID10, which provides fast rebuilds if a drive fails, elevator
seeks (elevator algorithm) and load balancing between the pairs of drives.
And anyway, today you can get (at Amazon) much large drives such as WD
Purple Surveillance 8TB for US$208, or 12TB for $330 (compare that to your
12 drives @ 1TB below).

Tony Patti
crypto at glassblower.info