It worth to separate if you have faster drive, let’s say a write optimized nvme or optane drive. Istvan Szabo Senior Infrastructure Engineer --------------------------------------------------- Agoda Services Co., Ltd. e: istvan.szabo@xxxxxxxxx<mailto:istvan.szabo@xxxxxxxxx> --------------------------------------------------- On 2021. Dec 5., at 10:38, Md. Hejbul Tawhid MUNNA <munnaeebd@xxxxxxxxx> wrote: Email received from the internet. If in doubt, don't click any link nor open any attachment ! ________________________________ Dear Zakhar and Anthony, Thank you for your valuable feedback. Maybe, my information was wrong. after creating VG and LV our command was below to create osd for SATA-HDD ceph-deploy osd create --bluestore --data ceph-block-2/block-2 --block-db ceph-db-0/db-2 cephosd1 So, we are using separate block db. Currently we are planning to install new OSD with SATA-SSD. so block-db will not be required. we will use device-class and different replicated_rule $ceph osd crush set-device-class SSD osd.10 $ceph osd crush rule create-replicated replicated_rule_SSD default host SSD Are we good to go? Regards, Munna On Sun, Dec 5, 2021 at 3:09 PM Zakhar Kirpichenko <zakhar@xxxxxxxxx> wrote: Hi! If you use SSDs for OSDs, there's no real benefit from putting DB/WAL on a separate drive. Best regards, Z On Sun, Dec 5, 2021 at 10:15 AM Md. Hejbul Tawhid MUNNA < munnaeebd@xxxxxxxxx> wrote: Hi, We are running openstack cloud with backend ceph storage. Currently we have only HDD storage in our ceph cluster. Now we are planning to add new server and osd with SSD disk. Currently we are using separate SSD disk for journal disk. Now if we install new OSD with SSD disk, do we need separate SSD disk for the journal? What will be the best approach. Regards, Munna _______________________________________________ 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
 |