I am using Elastic Search v5.2.2.
I have a single node cluster with 3 primary shards & 0 replica shard.
Total documents are around 10 million with 5 GB of data.
I have disabled scoring and most of the fields in my index are not_analyzed.
I am also using the pre-load feature where I load all the files (index.store.preload : ["*"]) before beginning the search operations.
Currently, I have 128 GB RAM and 1.5 TB of hard disk....Heap size is set to 8 GB.
Inspite of having such a small amount of data and a high performance machine, my CPU spikes to 100% and performance is not as expected.
What might be the reason for such slow performance?
And what can be interpreted from the following hot thread output? I can see Scorers in the output even though I have disabled scoring.