Sizing recommendations

[mmuru]

The use case is concurrent (100 users) running the query a delta lake table (>100M rows) via Kyuubi REST APIs. I would like to ask Kyuubi team to provide a guidance and recommendation sizing requirements running Kyuubi in K8S.

Using Kyuubi server (1.6.0), Spark engine (3.2.1) and delta lake 2.0.0

Clients -> backend service -> Kyuubi server -> Spark engine -> Delta table

Please, let me know if any other information needed. Highly appreciated your help.

@pan3793 @turboFei