Hey Eric,

The product doesn't have that functionality built in, but we have this:
http://docs.neo4j.org/chunked/stable/ha-haproxy.html#_cache_based_sharding_with_haproxy

Another recommendation is to direct writes to master and reads to slaves,
and we have this:
http://docs.neo4j.org/chunked/stable/ha-haproxy.html#_optimizing_for_reads_and_writes


Hope that helps?

 - Lasse





On Mon, Jun 9, 2014 at 9:28 PM, Eric Olson <[email protected]> wrote:

> I am playing with a personal, enterprise license of v2.1.1. I have set up
> a HA cluster (3 instances) on my local machine. The cluster is up and
> running as verified through the webadmin console and writes to master and
> slaves are populating through out the cluster as expected. Basically, it
> all works!
>
> What I am unsure of is any advanced functionality for how reads and the
> cache are handled in the cluster. I looked up cache-sharding and feel like
> I understand all that - having the server determine which node in the
> cluster to hit. BUT, is there an automatic or built-in functionality for
> distributing the load throughout the cluster?
>
> Another way to ask would be, does the cluster expose an endpoint (like a
> port on a server) that will handle the distribution of queries
> automagically? I notice when I hit the Neo4j instance running at :7474 that
> the cache size, hits and misses update while the other two instances have
> no activity. Obviously in this way I am only utilizing the one instance.
>
> Any guidance would be helpful and appreciated :)
>
> --
> You received this message because you are subscribed to the Google Groups
> "Neo4j" 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 
"Neo4j" 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