Hi Anthony-san, Thank you for your advice. I confirm my settings of my ceph cluster. Autoscaler mode is on, so I had thought it's the best PGs. But the autoscaler feature doesn't affect OSD's PGs. It's just for PG_NUM in storage pools. Is that right? Regards, -- Mitsumasa KONDO 2024年4月15日(月) 22:58 Anthony D'Atri <anthony.datri@xxxxxxxxx>: > If you're using SATA/SAS SSDs I would aim for 150-200 PGs per OSD as shown > by `ceph osd df`. > If NVMe, 200-300 unless you're starved for RAM. > > > > On Apr 15, 2024, at 07:07, Mitsumasa KONDO <kondo.mitsumasa@xxxxxxxxx> > wrote: > > > > Hi Menguy-san, > > > > Thank you for your reply. Users who use large IO with tiny volumes are a > > nuisance to cloud providers. > > > > I confirmed my ceph cluster with 40 SSDs. Each OSD on 1TB SSD has about > 50 > > placement groups in my cluster. Therefore, each PG has approximately 20GB > > of space. > > If we create a small 8GB volume, I had a feeling it wouldn't be > distributed > > well, but it will be distributed well. > > > > Regards, > > -- > > Mitsumasa KONDO > > > > 2024年4月15日(月) 15:29 Etienne Menguy <etienne.menguy@xxxxxxxxxxx>: > > > >> Hi, > >> > >> Volume size doesn't affect performance, cloud providers apply a limit to > >> ensure they can deliver expected performances to all their customers. > >> > >> Étienne > >> ------------------------------ > >> *From:* Mitsumasa KONDO <kondo.mitsumasa@xxxxxxxxx> > >> *Sent:* Monday, 15 April 2024 06:06 > >> *To:* ceph-users@xxxxxxx <ceph-users@xxxxxxx> > >> *Subject:* Performance of volume size, not a block size > >> > >> [Some people who received this message don't often get email from > >> kondo.mitsumasa@xxxxxxxxx. Learn why this is important at > >> https://aka.ms/LearnAboutSenderIdentification ] > >> > >> Hi, > >> > >> In AWS EBS gp3, AWS says that small volume size cannot achieve best > >> performance. I think it's a feature or tendency of general > >> distributed storages including Ceph. Is that right in Ceph block > storage? I > >> read many docs on ceph community. I never heard of Ceph storage. > >> > >> > >> > https://can01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdocs.aws.amazon.com%2Febs%2Flatest%2Fuserguide%2Fgeneral-purpose.html&data=05%7C02%7Cetienne.menguy%40ubisoft.com%7C3076825a4d2a4897074208dc5d017852%7Ce01bd386fa514210a2a429e5ab6f7ab1%7C0%7C0%7C638487508098942744%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=wOQKqG41uccTbyNHDIps62ojcTFBZYlyxxp3TzccsJI%3D&reserved=0 > >> <https://docs.aws.amazon.com/ebs/latest/userguide/general-purpose.html> > >> > >> Regard, > >> -- > >> Mitsumasa KONDO > >> _______________________________________________ > >> ceph-users mailing list -- ceph-users@xxxxxxx > >> To unsubscribe send an email to ceph-users-leave@xxxxxxx > >> > > _______________________________________________ > > ceph-users mailing list -- ceph-users@xxxxxxx > > To unsubscribe send an email to ceph-users-leave@xxxxxxx > > _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |