Open source platform for stress free deployments, measured impact, and smarter decisions.

GrowthBook Users

Hi GB Team,
We run experiment analysis with ~5 metrics in Clickhouse, and we have out of memory errors.
It seems that multiple queries are being executed at the same time. Is there a way to run these queries sequentially, and not in parallel? This way we think that there will be enough memory to handle each one of them.

You can use fact tables which run as one query

is there a missing index? how much data are you scanning?

```+--------------------------------+-----------+----------+-----------------+
|table                           |total_rows |size      |size_uncompressed|
+--------------------------------+-----------+----------+-----------------+
|12go.visit_log_ex               |570134337  |64.47 GiB |274.93 GiB       |
|12go.event_logs                 |703375698  |22.39 GiB |411.06 GiB       |
|12go.mv_log_checkout            |240325231  |20.17 GiB |238.48 GiB       |
|12go.experiment_log_new         |565249790  |17.79 GiB |118.68 GiB```

what is running out of memory? the query?

```500: Code: 241. DB::Exception: Memory limit (total) exceeded: would use 50.89 GiB (attempt to allocate chunk of 12582912 bytes), maximum: 49.43 GiB. OvercommitTracker decision: Query was selected to stop by OvercommitTracker.: While executing JoiningTransform. (MEMORY_LIMIT_EXCEEDED) (version 23.7.4.5 (official build))```
While calculating one of the metrics for analysis

We're trying to overcome it now by queuing up the requests at HAproxy level

I think the queries might need some optimizing

I’m being told by our DB guys that clickhouse can’t efficiently execute this kind of queries 

ya, we know that clickhouse doesn't have great inbuilt optimizers

we know some folks there, if you need some help optimizing

Now I’m out of office. I’ll provide the details on Monday, ok? Many thanks for the response 

Also the query for the Health tab is a heavy one 