[ClusterLabs] corosync/pacemaker on ~100 nodes cluser

[Kai Dupke]

On 08/23/2016 06:26 PM, Radoslaw Garbacz wrote:
> I would like to ask for settings (and hardware requirements) to have
> corosync/pacemaker running on about 100 nodes cluster.

Not bad ...

For the SUSE HA offering we limit support to 32 nodes because after this
the timing gets so critical that we can't make a support statement anymore.

We do have customers with special testing / setups running ~40 nodes,
but that really depends on the workload and resources you have to configure.

Looks like you've found this, too.

pacemaker_remote might be a chance to get more nodes, depending on the
services you're running.

I wonder if not a cloud solution (with some HA embedded) or a break up
in smaller chunks is beneficial.

It's not only about the technical capabilities but the mental
capabilities of the admins limiting the reliability of such a huge cluster.

Just in case, there are HPC clusters with many more nodes, but these
address a total different use case.

regards,
Kai Dupke
Senior Product Manager, HA & HPC
-- 
Sell not virtue to purchase wealth, nor liberty to purchase power.
Phone:  +49-(0)5102-9310828     Mail: kdupke at suse.com
Mobile: +49-(0)173-5876766      WWW:  www.suse.com

SUSE Linux GmbH - Maxfeldstr. 5 - 90409 Nuernberg (Germany)
GF:Felix Imendörffer,Jane Smithard,Graham Norton,HRB 21284 (AG Nürnberg)