Bringing together the Apache Cassandra experts from the community and DataStax.
Want to learn? Have a question? Want to share your expertise? You are in the right place!
Not sure where to begin? Getting Started
I am trying to figure out whether Cassandra uses `threads per connection` or `thread per operation`. I've observed `tpstats` output while workload is running but it is not giving any clear picture about it from the thread-pool. I've seen one process thread active all time when workload is running which is also not helping much to give an idea whether it is per connection base or per operation base.
Is there a utility / parameter or any way to check how threads are spawned, apart from `tpstats`?
On the client side, the driver by default opens a single connection to each node in the cluster and sends asynchronous "streams" using the same connection where each request is tagged with a stream ID.
On the server-side, each client request is handled by a thread which maps to Native-Transport-Requests in the nodetool tpstats output as you already discovered. Cheers!
7 People are following this question.
