|
All, I wish to experiment with erasure-coded pools in Ceph.
I’ve got some questions: 1.
Is FIREFLY a reasonable release to be using to try EC pools?
When I look at various bits of development info, it appears that the work is complete in FIREFLY, but I thought I’d askJ 2.
It looks, in FIREFLY, as if not all I/O operations can be performed on EC pools.
I am trying to work with RBD clients, and I’ve run into some conflicting information… can
RBDs run on EC pools directly, or is a caching tier required?
a.
Assuming a cache tier is required, where might I read information on sizing the cache tier?
b.
Looking through the issues, it appears there are some race conditions (e.g.,
#9285)
for cache tiers in FIREFLY. Should I avoid cache tiers at this level?
At what level, if any, are these addressed (I don’t see commits in
#9285, for example)? 3.
When configuring EC pools, to specify the number of PGs, can I reasonably assume that I should use (K+M) instead of a replica count?
So, for example, if I have 24 OSDs, and my EC profile has K=8 and M=4, then I should specify 200 (i.e., (24*100)/12) placement groups? 4.
As I add OSDs, can I adjust the number of PGs? Thanks in advance… _______________________________________________________ Don Doerner Quantum Corporation The information contained in this transmission may be confidential. Any disclosure, copying, or further distribution of confidential information is not permitted unless such privilege is explicitly granted in writing by Quantum. Quantum reserves the right to have electronic communications, including email and attachments, sent across its networks filtered through anti virus and spam software programs and retain such messages in order to comply with applicable data security and retention requirements. Quantum is not responsible for the proper and complete transmission of the substance of this communication or for any delay in its receipt. |
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
