Oliver Tennert wrote:
Hello,
has anybody ever heard of RAID^n technology, developed and patented by a
company called Inostor (http://www.inostor.com)?
It is designed to offer protection against the simultaneous failure of an
arbitrary number of arbitrary disks.
The pdf files publicly available on their homepage states that the
algorithms used by them to implement an m+n RAID system (m data disks, n
redundancy disks) are NOT the well-known Reed-Solomon codes as e.g.
described in a well-known paper by J. Plank.
This paper is [1]. But AFAIK it describes only Vandermonde-based
Reed-Solomon codes, not Cauchy-based ones.
Does anybody have a notion how RAID^n works? Their algorithms used seem to
hold a regular US patent.
RAID^n is nothing more than FEC (Forward Error Correction). Any erasure
code can be used to implement FEC:
- Cauchy-based Reed-Solomon codes
- Vandermonde-based Reed-Solomon codes
- Low-Density Parity-Check codes [2] (eg: Tornado Codes...)
Tornado codes are patented by Digital Fountain [3] (RFC 3453 contains a
list of the patents). Maybe Inostor uses them ?
[1] http://www.cs.utk.edu/~plank/plank/papers/CS-96-332.html
[2] http://www.cs.utk.edu/~plank/plank/papers/CS-03-510.html
[3] http://www.digitalfountain.com
--
Marc Bevand
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
