Grafana Cloud vs Clickhouse for this application

I have a very general high level design question I want your advice on.

Let’s consider a stock trading dashboard application frontend, which has to:

  1. allow the user to query the price for a particular stock in a time range in the past
  2. real-time updates for prices of selected securities.

I have two design alternatives I am considering:

  1. Grafana cloud (Mimir for long term storage)
  2. Clickhouse.

The Clickhouse stack currently presents many advantages such as easily exporting the data to numpy for easy computations of indicators etc. on the backend.

Are there advantages with the Grafana stack at all that I should consider?

I would say you are comparing two different things. Grafana stack (multiple products such as Grafana, Mimir, Loki,…) with single product Clickhouse. It makes better sense to compare only a Mimir / Clickhouse.
I wouldn’t say they are competitors. They are built for different use cases - TSDB vs OLAP DB, so choose the right one for your use case.

What is the expected performance difference of using Mimir vs Clickhouse for the following two settings’

  • Range query for time series recently (i.e. in mem/disk)
  • Range query for time series far in the past (i.e. spill to S3?)