12+3 means that all writes touch every node. Notably, backfill / recovery are writes. I suggest that 6+3 would suffice wrt capacity but offer a better experience. When possible, there is benefit in having at least one more failure domain than replicas, otherwise every OSD down has an outsize impact on capacity.
> On Sep 7, 2022, at 5:22 PM, Bailey Allison <balli...@45drives.com> wrote: > > Just to add onto Danny, I think a K+M of 12+3 with setting the failure > domain at the host level would give you what you want? It would actually be > 80% useable rather than 75%, (12/15 = 80%, 3/15 = 20%) but you could lose 2 > full hosts and still have access to the cluster. > > The only downside is you could technically just lose for example 4 > individual OSDs across 4 hosts and if they share the same PG(s) you might > have not a very good time, but such is life with erasure coded pools. > > > -----Original Message----- > From: Danny Webb <danny.w...@thehutgroup.com> > Sent: September 7, 2022 11:51 AM > To: Christophe BAILLON <c...@20h.com>; ceph-users <ceph-users@ceph.io> > Subject: [ceph-users] Re: Advice to create a EC pool with 75% raw capacity > usable > > Seeing as you want to be able to lose 2 hosts and still have a writeable > cluster your minimum m is at least 3 (k+1 is required for a writeable > cluster usually). Your failure domain will have to be set to host and your > k+m can't be larger than your failure domain (so no number larger than 15 or > ideally smaller than that so the cluster could rebalance in event of hard > failure). I'm not entirely sure what the performance impact is of creating > an overly large stripe, you'll have to test that out. > ________________________________ > From: Christophe BAILLON <c...@20h.com> > Sent: 07 September 2022 15:25 > To: ceph-users <ceph-users@ceph.io> > Subject: [ceph-users] Advice to create a EC pool with 75% raw capacity > usable > > CAUTION: This email originates from outside THG > > Hello, > > I need advice on the creation of an EC profile and the associate crush rule, > for a cluster of 15 nodes, each with 12 x 18Tb disks with the objective of > being able to lose 2 hosts or 4 disks. > I would like to have the most space available, a 75% ratio would be ideal > > If you can give me some examples ou some good links, it will be nice > > Regards > > > _______________________________________________ > ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email > to ceph-users-le...@ceph.io > > > Danny Webb > Principal OpenStack Engineer > The Hut Group<http://www.thehutgroup.com/> > > Tel: > Email: danny.w...@thehutgroup.com<mailto:danny.w...@thehutgroup.com> > > For the purposes of this email, the "company" means The Hut Group Limited, a > company registered in England and Wales (company number 6539496) whose > registered office is at Fifth Floor, Voyager House, Chicago Avenue, > Manchester Airport, M90 3DQ and/or any of its respective subsidiaries. > > Confidentiality Notice > This e-mail is confidential and intended for the use of the named recipient > only. If you are not the intended recipient please notify us by telephone > immediately on +44(0)1606 811888 or return it to us by e-mail. Please then > delete it from your system and note that any use, dissemination, forwarding, > printing or copying is strictly prohibited. Any views or opinions are solely > those of the author and do not necessarily represent those of the company. > > Encryptions and Viruses > Please note that this e-mail and any attachments have not been encrypted. > They may therefore be liable to be compromised. Please also note that it is > your responsibility to scan this e-mail and any attachments for viruses. We > do not, to the extent permitted by law, accept any liability (whether in > contract, negligence or otherwise) for any virus infection and/or external > compromise of security and/or confidentiality in relation to transmissions > sent by e-mail. > > Monitoring > Activity and use of the company's systems is monitored to secure its > effective use and operation and for other lawful business purposes. > Communications using these systems will also be monitored and may be > recorded to secure effective use and operation and for other lawful business > purposes. > > hgvyjuv > _______________________________________________ > ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email > to ceph-users-le...@ceph.io > > _______________________________________________ > ceph-users mailing list -- ceph-users@ceph.io > To unsubscribe send an email to ceph-users-le...@ceph.io _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-le...@ceph.io
