On Tue, Oct 18, 2016 at 3:46 AM, Lars Marowsky-Bree <l...@suse.com> wrote:
> On 2016-10-18T00:06:57, Erick Perez - Quadrian Enterprises <
> epe...@quadrianweb.com> wrote:
> > Is EC in the roadmap for CEPH? Cant seem to find it. My question is
> > "all others" (Nutanix, Hypergrid) do EC storage for VMs as the default
> > of storage. It seems EC in ceph (as of Sept 2016) is considered by many
> > "experimental" unless is used for cold data.
> Right now (Jewel) you have to "cheat" and put a cache tier in front of
> an EC pool if you want to use it for block or direct data access (like
> CephFS). That actually works quite well, and you can then also use
> much faster storage media in front of your dense EC pool.
> The downside is you have to understand how to set some of the cache
> tiering tunables and be able to size it appropriately, and understand
> the impact of cache tiering on your cluster performance (e.g., the
> overhead of the additional traffic between the tiers). Yet, it works
> Partial overwrites for EC pools directly is on the roadmap for
> Kraken/Luminous, probably leveraging some of BlueStore's new features.
> If I understood Sage correctly, he considers that a release criteria for
> That'd allow us to directly access EC pools w/o the need for cache
> SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB
> 21284 (AG Nürnberg)
> "Experience is the name everyone gives to their mistakes." -- Oscar Wilde
> ceph-users mailing list
Thanks Lars, I was reading about bluestore last night. I will wait for a
Kraken release to test as much EC functionality as possible at least from a
non-programmer point of use. Point and click tunables for EC is something
that will for sure be needed at least for some of us that do not run a
gazillion VMs and Petas of object storage.
Hopefully if i develop more know-how in EC I can help with testing
predefined tunables that can be available for smaller installations to use,
like the ones i use (always in Hewlett packard hardware) at least in my
country we do not run white-box hardware outside a lab.
I guess my first goal will be to learn how to obtain proper meaningful
metrics when testing VM loads and object loads.
Thanks to all for the comments,
ceph-users mailing list