Hi,

I am running experiments with different kinds of workloads using Neo4j 
client-server architecture.
When I run the system with actions that query the database, my observed 
throughput is constant
throughout the experiment.  However, with a mix of read (query) and write 
(update to nodes and
edge creation/deletion) actions, the throughput starts high and drops after 
hundreds of seconds.
The drop is sharp from a thousand to a few hundred actions per second.  
This happens only when
my workload is multi-threaded.  (The throughput does not drop with one 
thread.)

Is there a good explanation for this?

-- 
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