Hi,

You may need to consider the latency between the az's, it may make it
difficult to get very high iops - I suspect that is the reason ebs is
replicated within a single AZ.

Have you any data that shows the latency between the az's?

Thanks

On Sat, 28 Jul 2018, 05:52 Mansoor Ahmed, <[email protected]> wrote:

> Hello,
>
> We are working on setting up Ceph on AWS i3 instances that have NVMe SSD
> as instance store to create our own EBS that spans multiple availability
> zones. We want to achieve better performance compared to EBS with
> provisioned IOPS.
>
> I thought it would be good to reach out to the community to see if any one
> has done this or if anyone would advice against it or any other advice that
> could be of help.
>
> Thank you for your help in advance.
>
> Regards
> Mansoor
> ᐧ
> _______________________________________________
> ceph-users mailing list
> [email protected]
> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
>
_______________________________________________
ceph-users mailing list
[email protected]
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

Reply via email to