On 9/29/23 13:28, rajani m wrote:
What could cause some nodes in a cluster to have high query latency
when compared to the rest?
It is possible that some of the nodes have either handled zero queries
since the last reboot, or that they have handled fewer queries than the
others, and as a result they have less of the index sitting in the OS
disk cache. When Solr actually has to go out to the disk to read index
data, it is MUCH slower than when it can simply read that data directly
from memory.
Even SSD storage, prized for its performance, is slower than main memory.
https://cwiki.apache.org/confluence/display/SOLR/SolrPerformanceProblems
Thanks,
Shawn
