blobd

• Scales to millions of random concurrent partial reads over trillions of small objects with constant low latency.
• Asynchronous replication and event streaming.
• Batch creation API for very high transfer and creation rates with many small objects.
• HTTP RESTful API with support for CORS, range requests, presigned URLs, and HTTP/2.

Design

• On-device configurable fixed-size hash map with linked list of objects on the heap. The entire device is mapped to memory.
• Optimised for reads, then creates, then deletes. There is no way to list objects.
• Create an object, then write its data in 16 MiB aligned and exact chunks, then commit it.
• Objects are immutable once committed. Versioning is currently not possible. An object replaces all other objects with the same key when it's committed (not created).
• Only the size is stored with an object. No other metadata is collected, and custom metadata cannot be set.
• The device must be under 256 TiB. Objects are limited to 1 TiB. The peak optimal amount of objects stored is around 280 trillion.
• Uncommitted objects may be deleted after 7 days. Space used by objects may not be immediately freed when the object is deleted.

Turbostore

This project used to be called Turbostore and was written entirely in C; you can still see the code here.