News for hackers. Stuff that matters.

riccomini

DD supports lattices that allow it to compute at multiple points in time simultaneously. As I understand it, DBSP limits time to one diff at a time. Lalith can correct me if I’m off base on this. :)
Great question! We started out with the design you described--WAL as L0. But we found that there's a bit of a tension between wanting to have L0 SSTs be larger (and having fewer of them) to reduce metadata size, while we wanted to keep WAL SSTs small and frequent (to reduce async/await latency).

Basically, we wanted to have WAL writes go on the order of milliseconds, but we wanted L0 SSTs to be larger since they actually service reads.

The architecture page has more detail if you haven't found it yet:

https://slatedb.io/docs/architecture

Show HN: SlateDB – An embedded storage engine built on object storage

https://github.com/slatedb/slatedb

SlateDB is an embedded storage engine built as a log-structured merge-tree. Unlike traditional LSM-tree storage engines, SlateDB writes data to object storage (S3, GCS, ABS, MinIO, Tigris, and so on). Leveraging object storage allows SlateDB to provide bottomless storage capacity, high durability, and easy replication. The trade-off is that object storage has a higher latency and higher API...

( Read More... | 7 comments | 6 points | on HN )

Older »