I've been reading the documentation, and attempting to put together a PoC 
based on the technology.  We have a graph with about 50M records, with only 
a few basic relationships between nodes (no walks more than two nodes in 
length).

After attempting to execute "select * from X limit 100", the system chewed 
up all available memory (I had 8GB allocated), and then later crashed.  The 
console recovered a connection (kudos for that), but it's clear that we're 
going to need more in-depth configuration information, or to cluster the 
data, splitting a class type over multiple hosts.

I'm having a hard time locating documentation or examples on how to best 
cluster.  I've been looking at the unit tests, but those appear to be 
designed within the constraint of a single machine.

Is there an example application / tutorial that demonstrates:

1.  How to set up clustering on each server
2.  How to configure clients to observe clustering, routing queries to the 
appropriate host

(1) Appears to be provided in the existing documentation.  I just have yet 
to work through it.  (2) is a question mark.  If there is sufficient 
documentation for this, and I've just missed it, I'd appreciate the links 
to relevant pages.  But, it would be nice to find one single tutorial that 
provided step-by-step guidance in setting up multiple hosts, and 
demonstrating how queries are routed.

Thanks in advance!

-jjk

-- 

--- 
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