Hi, Although much of this has been discussed at IRC, i thought it might me helpfull to post it at the ml so i can be found easily. In my Ceph cluster i was experiencing a performance problem, which after some tests was due to a number of OSD's which had some disk I/O problems. At the moment i have 4 OSD's which all get performance of roughly 55MB/sec (ceph osd bench), where the journal and data is on the same disk (seperate partition though). Now i'm seeing about 50MB/sec with the rados benchmarks, which seem fine, but as i have 4 OSD's i expected to get a performance of about 100MB/sec (since my replication level is at two). The question is, how does Ceph scale? Will a single writer thread be limited to the performance of one OSD and will i get a higher performance when multiple threads are started? Can a dev shed some light on this topic? For the reference, you can benchmark your OSD with: ceph osd tell 0 bench Make sure you have "ceph -w" running somewhere so you can see the result. -- Met vriendelijke groet, Wido den Hollander Hoofd Systeembeheer / CSO Telefoon Support Nederland: 0900 9633 (45 cpm) Telefoon Support België: 0900 70312 (45 cpm) Telefoon Direct: (+31) (0)20 50 60 104 Fax: +31 (0)20 50 60 111 E-mail: support@xxxxxxxxxxxx Website: http://www.pcextreme.nl Kennisbank: http://support.pcextreme.nl/ Netwerkstatus: http://nmc.pcextreme.nl -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
