Damien Picard <picard.dam...@gmail.com> wrote: > Currently we have 4 Solr nodes, with 12Gb memory (heap) ; the collections > are replicated (4 shards, 1 replica). > This query mostly returns a QTime=4 and it takes around 20ms on the client > side to get the result.
> We have to handle around 200 simultaneous connections. You are probably experiencing congestion. JMeter can visualize throughput. Try experimenting with 10 to 100 concurrent threads in increments of 10 threads and look at throughput underway. My guess is that throughput will rise as you increase threads, until some point after which it will fall again as the Solrs exceeds their peak performance point. You might end up getting better performance by rate-limiting outside of SolrCloud. Also, what does 200 simultaneous connections mean? Is that 200 requests per second? - Toke Eskildsen
