Hi guys,

When running distributed, the existent clusters are assigned to the
servers, so no other clusters are created per servers, unless the number of
servers is > than existent clusters. In releases <v2.2 one cluster per
server was always created, so in the worst case scenario with 10 nodes and
8 cores, you could end up with 17 clusters per class.

In v2.2 they would be just 10 per class.

Furthermore you could decide to have only 3 master servers and 100 replica
only servers. Those servers don't create any additional clusters because
they are read only.

I hope now it's more clear.


Best Regards,

Luca Garulli
Founder & CEO
OrientDB <http://orientdb.com/>


On 25 April 2016 at 21:47, 'scott molinari' via OrientDB <
[email protected]> wrote:

> Good point. I wonder if the cpu cores = # of clusters setting also counts
> in a distributed system. It wouldn't seem to make too much sense, because
> "over-chunking" the data can also cause performance issues when querying.
>
> Scott
>
> --
>
> ---
> You received this message because you are subscribed to the Google Groups
> "OrientDB" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> For more options, visit https://groups.google.com/d/optout.
>

-- 

--- 
You received this message because you are subscribed to the Google Groups 
"OrientDB" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to