Carcinisation is the term given to the fact that every non-crab crustacean eventually evolves a crab body. Everything, eventually, becomes a crab... Any project that works with state in the cloud (database, streaming, observability etc), will eventually use blob storage as it's durable storage layer and just be a fancy API layer in front of it. Everything, eventually, becomes S3... I am watching the carcinisation of the database before my very eyes. The following "datastores" all are backed by S3 (blob) as their primary durability layer. This list was last updated 2nd October, 2024 - [Turso](https://turso.tech) - [Neon](https://neon.tech/docs/introduction/architecture-overview) - [Datastax Astra](https://docs.datastax.com/en/astra-serverless/docs/astra-apache-faq.html#_how_does_the_control_plane_work_with_astra_db) - [AWS Opensearch](https://aws.amazon.com/blogs/big-data/amazon-opensearch-serverless-is-now-generally-available/) - [Confluent Cloud](https://www.confluent.io/blog/cloud-native-data-streaming-kafka-engine/) - [WarpStream](https://www.warpstream.com) - [Snowflake](https://docs.snowflake.com/en/user-guide/intro-key-concepts#database-storage) - [Pinecone](https://www.pinecone.io/blog/serverless/) - [Honeycomb](https://x.com/mipsytipsy/status/1674140269632188416) - [Clickhouse](https://clickhouse.com/blog/building-clickhouse-cloud-from-scratch-in-a-year) - [AutoMQ](https://github.com/AutoMQ/automq)