-1C - Erasure Encoding as a Storage Backend¶

xiaoxi    I cannot see anyone & any words,it's all blank    09:00
loicd    sjust: \o    09:00
loicd    that's what the summit is for, right ? ;-)    09:01
*** rturk is now known as rturk-away    09:01
*** rturk-away is now known as rturk    09:01
loicd    http://wiki.ceph.com/01Planning/02Blueprints/Dumpling/Erasure_encoding_as_a_storage_backend    09:01
loicd    http://pad.ceph.com/p/Erasure_encoding_as_a_storage_backend    09:02
*** sjustlaptop has joined #ceph-summit1    09:02
sagewk    anybody else want to join the hangout?    09:03
nwl    in future, i wonder if it's better to keep the blueprint out of the etherpad    09:04
nwl    so one can just see the changes/comments etc    09:04
rturk    the etherpad only contains the work items    09:04
rturk    but ya, maybe it should just be for notes?    09:04
nwl    rturk: yeah, so the notes can be then fed back into the blueprint, redmine etc    09:05
rturk    (ideas welcome -> community@ceph.com, now or anytime!)    09:05
saras    have you look at http://www.opencompute.org/projects/storage/    09:05
*** liwang has joined #ceph-summit1    09:06
dmick    saras: I have not; did you mean relative to the management API work specifically, or in general?    09:09
nwl    saras: sage is involved in the project    09:09
saras    thanks    09:10
nwl    saras: we are not doing anything very actively though (as Inktank)    09:10
saras    can you guy setup lower thirds    09:11
saras    please    09:11
rturk    lost your audio, loicd    09:12
yehuda_hm    in practice an osd should not become full, unless something is misconfigured, or the entire cluster filled up    09:12
loicd    damned    09:13
loicd    ahahah    09:13
loicd    I'm back    09:13
saras    append would be great    09:13
*** maclin has joined #ceph-summit1    09:16
hskinner    erasure encoding is of course most valuable with geo-dispersment    09:17
sagewk    hskinner: true    09:17
sagewk    it's also not latency-sensitive in this use case...    09:17
hskinner    yes, ability to migrate to optimal region locality would be bonus    09:18
ksp    erasure encoding is also valuable in metro cluster/distance; e.g. EMC Isilon OneFS is using such algorithms    09:21
saras    is their paper that i can read about erasure encoding    09:22
dmick    saras: a bunch of resources on http://wiki.ceph.com/01Planning/02Blueprints/Dumpling/Erasure_encoding_as_a_storage_backend    09:24
saras    dmick: thanks    09:24
*** michael__ has joined #ceph-summit1    09:24
*** michael__ has quit IRC    09:24
*** yanzheng has quit IRC    09:26
*** michael_dreamhost has joined #ceph-summit1    09:26
ksp    http://en.wikipedia.org/wiki/Erasure_code    09:27
*** yanzheng has joined #ceph-summit1    09:27
saras    pg to be  clear    09:28