Hi all,

I'm looking to get a general overview of the performance tuning knobs and
recommendations that exist for Atlas. I looked through the docs, but
haven't seen anything.

In particular, are there any recommended HBase/Solr tuning
recommendations/pitfalls that've been seen with Atlas' use of JanusGraph?

I know there's some performance debugging information, but what's the best
way to identify bottlenecks in the overall flow (i.e. what needs to be
scaled out / tuned first from Kafka/HBase/Solr)? Is there anything that
needs to be done to avoid negative impact from reads vs writes? Are there
any JanusGraph settings that should be tweaked to fit Atlas' use case?

Finally, are there any benchmarks of reasonable performance against given
cluster configurations?

Thanks,
Justin

Reply via email to