Heavy bulk request on single index is slow

Elastic Stack Elasticsearch
1

Hi guys,

I am experiencing some weird thing. I am having 20 indexes. So when i do bulk insert/update on all index, The thread pool shows me

curl -XGET http://localhost:9200/_cat/thread_pool
localhost bulk                24 7881 0

And whenever i am heaving bulk request on single index it showing me

curl -XGET http://localhost:9200/_cat/thread_pool
localhost bulk                24 350 0

Both queue size is different. Any idea how i can scale ?

version : 6.2
jvm : 30g
OS : centos7

thread_pool.index.size : 24
thread_pool.index.queue_size : 10000

Thanks

2

I would recommend you have a look at this blog post as I believe it describes how this works.

3

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.