Re: CEPHFS with Erasure Coded Pool for Data and Replicated Pool for Meta Data

[Mohamed Pakkeer <mdfakkeer@xxxxxxxxx>]

Hi Greg,

We are planning to create 3 PB EC based storage cluster initially. what would be the recommended hardware configuration for creating caching pool? How many nodes will cache pool require to cater the 3 PB storage cluster? What is the size and network connectivity of each node?

-- Mohammed Pakkeer

On Wed, Jan 21, 2015 at 11:25 AM, Gregory Farnum <greg@xxxxxxxxxxx> wrote:

You can run CephFS with a caching pool that is backed by an EC pool,
but you can't use just an EC pool for either of them. There are
currently no plans to develop direct EC support; we have some ideas
but the RADOS EC interface is way more limited than the replicated
one, and we have a lot of other things we'd like to get right first.
:)
-Greg

On Tue, Jan 20, 2015 at 9:53 PM, Mohamed Pakkeer <mdfakkeer@xxxxxxxxx> wrote:
> Hi Greg,
>
> Thanks for your reply. Can we have mixed pools( EC and replicated) for
> CephFS data and metadata or we have to use  anyone pool( EC or Replicated)
> for creating CephFS? Also we would like to know, when will the production
> release of CephFS happen with erasure coded pool ? We are ready to test
> peta-byte scale CephFS cluster with erasure coded pool.
>
>
> -Mohammed Pakkeer
>
> On Wed, Jan 21, 2015 at 9:11 AM, Gregory Farnum <greg@xxxxxxxxxxx> wrote:
>>
>> On Tue, Jan 20, 2015 at 5:48 AM, Mohamed Pakkeer <mdfakkeer@xxxxxxxxx>
>> wrote:
>> >
>> > Hi all,
>> >
>> > We are trying to create 2 PB scale Ceph storage cluster for file system
>> > access using erasure coded profiles in giant release. Can we create
>> > Erasure
>> > coded pool (k+m = 10 +3) for data and replicated (4 replicas) pool for
>> > metadata for creating CEPHFS? What are the pros and cons of using two
>> > different pools to create CEPHFS ?
>>
>> It's standard to use separate pools. Unfortunately you can't use EC
>> pools for CephFS right now.
>> -Greg
>
>
>
>
>

_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com