[Gluster-users] GlusterFS for virtualization + RAID10 (HDD)

[Strahil Nikolov]

I would pick the approach of 1 RAID1 disk = 1 brick with the combination of LACP (hashing over IP+port ).
The fuse client's connection to the relevant brick will be using a different NIC port (cause LACP hashing based on IP+port will pick a different tcp port) and this way you will load balance and consume more bandwidth.

Also consider using cluster.choose-local which will prefer local bricks for reading.



Best Regards,
Strahil Nikolov






В петък, 16 април 2021 г., 21:28:35 ч. Гринуич+3, Gilberto Ferreira <gilberto.nunes32 at gmail.com> написа: 





Hi there

I am about to deploy a gluster server with Proxmox VE.
Both servers have 4 SAS disks configured as RAID10.
RAID10 is ok? Network is 4 1GB NIC. I am thinking of using some kind
of bond with 3 1G NICS. What about it?
Thanks for any advice.
---
Gilberto Nunes Ferreira
(47) 99676-7530 - Whatsapp / Telegram
________



Community Meeting Calendar:

Schedule -
Every 2nd and 4th Tuesday at 14:30 IST / 09:00 UTC
Bridge: https://meet.google.com/cpu-eiue-hvk
Gluster-users mailing list
Gluster-users at gluster.org
https://lists.gluster.org/mailman/listinfo/gluster-users