Snowflake is easily the best alternative. It's a little expensive, but it just works.

Slowflake isn’t really an alternative if someone is switching due to costs. It’s generally more expensive for the same workloads.

I could never understand how to go about guesstimating what snowflake would cost, their pricing abstraction feels like lunacy as an outsider.

That is because they figure out what you will pay and that is what they charge. It is completely disconnected from the costs involved.

They're the Oracle of cloud databases.

Great product, pricing models are pretty insane.

And despite the pricing, they're still losing money!

Snowflake is much slower than BigQuery in my experience.

For which workload?

My workload involves streaming around a terabyte per hour and creating cubes(varying aggs, count distinct) on that data once per hour in addition to daily, weekly, monthly. Snowflake frankly could not do it. The hourly cube took longer than an hour to finish.

A little expensive is an understatement. It's black magic to figure out how much it's going to cost. Our test instance of snowflake costs 10x production BigQuery costs, which stores an order of magnitude more data and queries even more

Athena and Redshift Spectrum are the best alternatives on the AWS side.

Buy the biggest server on Hetzner and install ClickHouse (/s?)

There's also ClickHouse Cloud now.

Even better, there are lots of ClickHouse clouds! ClickHouse Cloud, Altinity.Cloud, DoubleCloud, Aiven, TinyBird...plus many others. This is the best assurance for users that prices for ClickHouse will stay competitive over time.

Disclaimer: I work at Altinity.

Snowflake seems most similar, plus you would get to stay on GCP.

Export your data via avro files and try using Athena

Ideally you’d used Parquet or ORC if querying from Athena/Trino/Presto. Since they are columnar formats you will enjoy considerably faster queues and lower query costs for most query patterns other than “select *…” since the query engine can just retrieve the columns needed for your query instead of the entire row.

100 percent correct.

Ideally you will also partition the data, so your queries can use partitions. Athena charges $5 per terabyte of data scanned, so it's important to get this right.

May be a slightly different use case but moving from BigQuery to managed Clickhouse on clickhouse.com has been pretty great. Easy data injection, super fast querying, lots of control over the environment, and a nice SQL console for quickly looking at the data.

We at https://www.gumlet.com use bigquery a lot, but now it seems we will have to move to click house. Can’t trust google with anything. First Kubernetes price intro, then multi region storage price increase along with inter region price introduction and now this.

I would trust AWS but not Google.