Hi Loic, I think it is good to have the flexibility to include the global parity stripes in local subgroups or as in the picture to make a subgroup out of the global parities and compute a local paritiy for them. Although with the chosen parameters these local subgroups are not symmetric (5,5 and 4 chunks per group) but e.g. for 8+4 one would have 3 symmetric local groups as drawn and if you project is on a three data center setup you would like to have for each center a local parity. But wasn't this already possible with your prototype implementation? Cheers Andreas. ________________________________________ From: Loic Dachary [loic@xxxxxxxxxxx] Sent: 09 May 2014 17:35 To: Andreas Joachim Peters Cc: Ceph Development Subject: Implied parity and erasure code Hi Andreas, The "implied parity bloc" mentionned page 4 of http://anrg.usc.edu/~maheswaran/Xorbas.pdf is something we've not discussed in the context of the implementation of the pyramid erasure code plugin. I'm not sure if it would be useful to have. Do you have an opinion ? Cheers -- Loïc Dachary, Artisan Logiciel Libre -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
