Re: Is it a bad Idea to build a Ceph Cluster over different Data Centers?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Den tis 10 aug. 2021 kl 21:59 skrev Ralph Soika <ralph.soika@xxxxxxxxx>:
> communicate via public IPs. I did some tests and - not surprising - the
> network performance is going down to 500Mbit/s between two nodes in two
> different data centers.
> My questions are:
>
>  1. Is 500Mbit/s to slow to run a ceph cluster?

This means the optimal read/write speed would be 50MB/s at most if the
client is not at the same DC as the primary OSD it wants to write to,
which is ~1/4 of what you get from a decent spin drive.

In case you have repl=3, OSDs at both sites and at least some data
needs to get copied back again over the link, the speed might drop
again leaving you with something like 25MB/s at most. Regardless of if
your dual DC set up is perfect in all other regards, this cap on speed
would probably be noticeable to all clients.

-- 
May the most significant bit of your life be positive.
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx



[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux