Possible Use Cases

[joaoh82]

These are some of the paths we have been talking about taking the project. One does not necessarily exclude the other, but we possibly will have to prioritize.

Out of each one of these, I plan to drill down until we have a viable plan.

1. Embedded DB that can run inside WASM, with in memory storage.

• Research best way to store data in a some what persistent manner.

2. Embedded DB locally or on a server, and backed by a disk data store like sled, and taking advantage of the concurrency features of Rust.

3. Embedded DB to use inside unit tests compatible with multiple SQL Dialects and features.

• One of the use case for an embedded database is to mock a 'bigger' database like MySQL or Postgres.

4. Use in self-contained micro services.

• Using a SQLite store would be 'too small' but orchestrating an external DB would be hard to manage. Often the choice is to pack a MySQL or Postgres within a docker image, but then it is not quite 'micro'. There is an opportunity for SQLrite to become the critical component of a self-contained micro service stack, together with other crates in the Rust ecosystem.

[tyt2y3]

'Embedded DB that can run inside WASM'. We can have a mode to load / store from a file format.
At startup, we load the database file from disk and work in memory.
At shutdown, we can dump the data back to the file.
Just that there will be no Atomic & Durable guarantee.

[tyt2y3]

For 2 & 4, I intended to mean the same. Because we are not going to have a 'wire protocol' for SQLrite, applications must link with the DB as one binary.
We can also provide bindings to other languages as SQLite do, but then they are also usually linked statically to the language runtime.
Or do you intend to have a wire protocol?

[joaoh82]

I see what you mean. You are totally right. I think the idea here would be provide bindings to other languages following the SQLite model. This way we don't have to worry about a wire protocol, which definitely complicates things.

So basically the idea would be to have a database embedded in a binary, but that would be able to handle not only multiple readers but also multiple writers, which today is a limitation for SQLite.

[joaoh82]

Merging 2 and 4 paths takes us here now:

1. Multi-dialect Embedded SQL Database.
2. Self-container read-write concurrent safe embedded DB
3. Embedded DB compiled to WASM. (Storage model TBD)