How messy is #terraform with #gcp?

I'm trying to make a system where #rust worker ingests data of HTTP requests via #cloudrun, and passes them into #bigtable, which uses further ingestion recipe to export the data into #bigquery

I have tried to make a complete terraform declaration for this, but got into permission issues, then I have tried to make a system that generates all the artefacts like service accounts and docker image and then refers to those from terraform builds, but I almost don't see a value of doing it like that.

Does anyone have an example of #cloudrun #cdc? I am new to this and I feel really slow.

Help needed! I'm writing a media theory book chapter about Hadoop, BigTable, ACID, BASE and CAP for a non-technical audience. I'm looking for an _expert_ and practitioner from the field to doublecheck the 11 pages I wrote. Thank you for boosting. #phd #academia #Hadoop, #BigTable

"#Bigtable [...] had these internal control-plane entities (as part of the implementation) called tablet servers, which had large indexes, and at some point they became a scaling bottleneck. So the Bigtable engineers scratched their heads over how to make it scale, and realized that they could replace the tablet servers with Bigtables to unblock the scaling. So Bigtable is part of its own implementation. It’s Bigtables all the way down" https://steve-yegge.medium.com/dear-google-cloud-your-deprecation-policy-is-killing-you-ee7525dc05dc #Google