Project

General

Profile

-1C - Erasure Encoding as a Storage Backend » History » Version 1

Jessica Mack, 06/22/2015 05:44 AM

1 1 Jessica Mack
h1. -1C - Erasure Encoding as a Storage Backend
2 1 Jessica Mack
3 1 Jessica Mack
<pre>
4 1 Jessica Mack
xiaoxi	I cannot see anyone & any words,it's all blank	09:00
5 1 Jessica Mack
loicd	sjust: \o	09:00
6 1 Jessica Mack
loicd	that's what the summit is for, right ? ;-)	09:01
7 1 Jessica Mack
*** rturk is now known as rturk-away	09:01
8 1 Jessica Mack
*** rturk-away is now known as rturk	09:01
9 1 Jessica Mack
loicd	http://wiki.ceph.com/01Planning/02Blueprints/Dumpling/Erasure_encoding_as_a_storage_backend	09:01
10 1 Jessica Mack
loicd	http://pad.ceph.com/p/Erasure_encoding_as_a_storage_backend	09:02
11 1 Jessica Mack
*** sjustlaptop has joined #ceph-summit1	09:02
12 1 Jessica Mack
sagewk	anybody else want to join the hangout?	09:03
13 1 Jessica Mack
nwl	in future, i wonder if it's better to keep the blueprint out of the etherpad	09:04
14 1 Jessica Mack
nwl	so one can just see the changes/comments etc	09:04
15 1 Jessica Mack
rturk	the etherpad only contains the work items	09:04
16 1 Jessica Mack
rturk	but ya, maybe it should just be for notes?	09:04
17 1 Jessica Mack
nwl	rturk: yeah, so the notes can be then fed back into the blueprint, redmine etc	09:05
18 1 Jessica Mack
rturk	(ideas welcome -> community@ceph.com, now or anytime!)	09:05
19 1 Jessica Mack
saras	have you look at http://www.opencompute.org/projects/storage/	09:05
20 1 Jessica Mack
*** liwang has joined #ceph-summit1	09:06
21 1 Jessica Mack
dmick	saras: I have not; did you mean relative to the management API work specifically, or in general?	09:09
22 1 Jessica Mack
nwl	saras: sage is involved in the project	09:09
23 1 Jessica Mack
saras	thanks	09:10
24 1 Jessica Mack
nwl	saras: we are not doing anything very actively though (as Inktank)	09:10
25 1 Jessica Mack
saras	can you guy setup lower thirds	09:11
26 1 Jessica Mack
saras	please	09:11
27 1 Jessica Mack
rturk	lost your audio, loicd	09:12
28 1 Jessica Mack
yehuda_hm	in practice an osd should not become full, unless something is misconfigured, or the entire cluster filled up	09:12
29 1 Jessica Mack
loicd	damned	09:13
30 1 Jessica Mack
loicd	ahahah	09:13
31 1 Jessica Mack
loicd	I'm back	09:13
32 1 Jessica Mack
saras	append would be great	09:13
33 1 Jessica Mack
*** maclin has joined #ceph-summit1	09:16
34 1 Jessica Mack
hskinner	erasure encoding is of course most valuable with geo-dispersment	09:17
35 1 Jessica Mack
sagewk	hskinner: true	09:17
36 1 Jessica Mack
sagewk	it's also not latency-sensitive in this use case...	09:17
37 1 Jessica Mack
hskinner	yes, ability to migrate to optimal region locality would be bonus	09:18
38 1 Jessica Mack
ksp	erasure encoding is also valuable in metro cluster/distance; e.g. EMC Isilon OneFS is using such algorithms	09:21
39 1 Jessica Mack
saras	is their paper that i can read about erasure encoding	09:22
40 1 Jessica Mack
dmick	saras: a bunch of resources on http://wiki.ceph.com/01Planning/02Blueprints/Dumpling/Erasure_encoding_as_a_storage_backend	09:24
41 1 Jessica Mack
saras	dmick: thanks	09:24
42 1 Jessica Mack
*** michael__ has joined #ceph-summit1	09:24
43 1 Jessica Mack
*** michael__ has quit IRC	09:24
44 1 Jessica Mack
*** yanzheng has quit IRC	09:26
45 1 Jessica Mack
*** michael_dreamhost has joined #ceph-summit1	09:26
46 1 Jessica Mack
ksp	http://en.wikipedia.org/wiki/Erasure_code	09:27
47 1 Jessica Mack
*** yanzheng has joined #ceph-summit1	09:27
48 1 Jessica Mack
saras	pg to be  clear	09:28
49 1 Jessica Mack
</pre>