All of lore.kernel.org
 help / color / mirror / Atom feed
* coefficients
@ 2014-07-11 10:48 Koleos Fuscus
  2014-07-11 17:19 ` coefficients Loic Dachary
  0 siblings, 1 reply; 2+ messages in thread
From: Koleos Fuscus @ 2014-07-11 10:48 UTC (permalink / raw)
  To: Loic Dachary; +Cc: ceph-devel@vger.kernel.org

Hi Loic,

If you still need to calculate more coefficients there is an
explanation of how to construct the coding equations in the paper
Erasure coding in Windows Azure Storage (unless you will keep your LRC
implementation simple and do only the LRC (10,6,5) ). I undertand that
the codes LRC in the mentioned paper are not exactly the same but at
least the paper explains the coefficients. I think in Azure they used
functional repair instead of exact repair as you wish.

koleosfuscus

________________________________________________________________
"My reply is: the software has no known bugs, therefore it has not
been updated."
Wietse Venema

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: coefficients
  2014-07-11 10:48 coefficients Koleos Fuscus
@ 2014-07-11 17:19 ` Loic Dachary
  0 siblings, 0 replies; 2+ messages in thread
From: Loic Dachary @ 2014-07-11 17:19 UTC (permalink / raw)
  To: Koleos Fuscus; +Cc: ceph-devel@vger.kernel.org

[-- Attachment #1: Type: text/plain, Size: 1095 bytes --]

Hi koleosfuscus,

Thanks for the pointer, it is very helpful. I'm inclined to keep the simpler approach for the first implementation while exploring and learning more about the smarter / more complex approach. This may change if someone willing to help comes on board in the meantime ;-)

Cheers

On 11/07/2014 13:48, Koleos Fuscus wrote:
> Hi Loic,
> 
> If you still need to calculate more coefficients there is an
> explanation of how to construct the coding equations in the paper
> Erasure coding in Windows Azure Storage (unless you will keep your LRC
> implementation simple and do only the LRC (10,6,5) ). I undertand that
> the codes LRC in the mentioned paper are not exactly the same but at
> least the paper explains the coefficients. I think in Azure they used
> functional repair instead of exact repair as you wish.
> 
> koleosfuscus
> 
> ________________________________________________________________
> "My reply is: the software has no known bugs, therefore it has not
> been updated."
> Wietse Venema
> 

-- 
Loïc Dachary, Artisan Logiciel Libre


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 263 bytes --]

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2014-07-11 17:19 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-07-11 10:48 coefficients Koleos Fuscus
2014-07-11 17:19 ` coefficients Loic Dachary

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.