SQL has two aggregation modes: facet and map_reduce. Facet uses the json facet API directly so SOLR-7452 would apply if it hasn't been resolved yet. map_reduce always gives accurate results regardless of the cardinality but is slower. To increase performance using map_reduce you need to increase the size of the cluster (workers, shards, replicas).
Joel Bernstein http://joelsolr.blogspot.com/ On Mon, Apr 24, 2017 at 5:09 AM, Furkan KAMACI <furkankam...@gmail.com> wrote: > Hi, > > As you know that json facet api returns inconsistent counts in cloud set up > (SOLR-7452). I would like to learn that is the situation same for Solr SQL > queries too? > > Kind Regards, > Furkan KAMACI >
