Good topic, I'd be interested also. One of the redhat document says 1GW / 50 OSD, but I think it is not a relevant formula. I had couple of time when the users doing something stupid and totally ddos down the hole cluster. What I've done added additional 4 rgw in each of the mon/mgr nodes where the gateway is running to sustain the super high load, so currently I'm using like 15 RGW behind a haproxy loadbalancer on 3 nodes.
Istvan Szabo Senior Infrastructure Engineer --------------------------------------------------- Agoda Services Co., Ltd. e: [email protected] --------------------------------------------------- -----Original Message----- From: [email protected] <[email protected]> Sent: Saturday, September 11, 2021 1:51 PM To: Eugen Block <[email protected]> Cc: ceph-users <[email protected]> Subject: [ceph-users] Re: How many concurrent users can be supported by a single Rados gateway Email received from the internet. If in doubt, don't click any link nor open any attachment ! ________________________________ Thanks for the suggestions. My viewpoints may be wrong, but i think stability is utmost for us, and an older version such as Luminous may be much well battle-field tested that recent ones. Unless there is some instatbilty or bug reports, I would still trust older versions. Just my own preference on which version takes my turst thanks a lot, Samuel [email protected] From: Eugen Block Date: 2021-09-10 17:21 To: huxiaoyu CC: ceph-users Subject: Re: [ceph-users] How many concurrent users can be supported by a single Rados gateway The first suggestion is to not use Luminous since it’s already EOL. We noticed major improvements in performance when upgrading from L to Nautilus, and N will also be EOL soon. Since there are some reports about performance degradation when upgrading to Pacific I would recommend to use Octopus. Zitat von [email protected]: > Dear Cephers， > > I am planning a Ceph Cluster (Lumninous 12.2.13) for hosting on-line > courses for one university. The data would mostly be video media and > thus 4+2 EC coded object store together with CivetWeb RADOS gateway > will be utilized. > > We plan to use 4 physical machines as Rados gateway solely, each with > 2x Intel 6226R CPU and 256 GB memory, for serving 8000 students > concurrently, of which each may incur 2x 2Mb/s video streams. > > Are these 4-machine Rados gateway a reasonable configuration for > 8000 users, or an overkill, or insufficient? > > Suggestions and comments are highly appreciated, > > best regards, > > Samuel > > > > [email protected] > _______________________________________________ > ceph-users mailing list -- [email protected] To unsubscribe send an > email to [email protected] _______________________________________________ ceph-users mailing list -- [email protected] To unsubscribe send an email to [email protected] _______________________________________________ ceph-users mailing list -- [email protected] To unsubscribe send an email to [email protected]
