fjall

*Fjall* _(Nordic: "Mountain")_ is a log-structured, embeddable key-value storage engine written in Rust. It features: - A thread-safe BTreeMap-like API - 100% safe & stable Rust - LSM-tree-based storage similar to `RocksDB` - Range & prefix searching with forward and reverse iteration - Multiple keyspaces (a.k.a. column families) with cross-keyspace atomic semantics - Built-in compression (default = `LZ4`) - Serializable transactions (optional) - Key-value separation for large blob use cases (optional) - Custom compaction filters to run custom logic during compactions (optional) - Automatic background maintenance It is not: - A standalone database server - A relational or wide-column database: it has no built-in notion of columns or query language ## Sponsors ## Basic usage ```bash cargo add fjall ``` ``` … ``` > [!TIP] > Like any typical key-value store, keys are stored in lexicographic order. > If you are storing integer keys (e.g. timeseries data), you should use the big endian form to have predictable ordering. ## Durability To support different kinds of workloads, Fjall is agnostic about the type of durability your application needs. After writing data (`insert`, `remove` or committing a write batch/transaction), you can choose to call [`Database::persist`](https://docs.rs/fjall/latest/fjall/struct.Database.html#method.persist) which takes a [`PersistMode`](https://docs.rs/fjall/latest/fjall/enum.PersistMode.html) parameter. By default, any operation will flush to OS buffers, but **not** to disk. This matches RocksDB's default durability. Also, when dropped, the database will try to persist the journal *to disk* synchronously. ## Multithreading, Async and Multiprocess > [!WARNING] > A single database may **not** be loaded in parallel from separate *processes*. Fjall is internally synchronized for multi-*threaded* access, so you can clone around the `Database` and `Keyspace`s as needed, without needing to lock yourself. For an async example, see the [`tokio`](https://github.com/fjall-rs/fjall/tree/main/examples/tokio) example. ## Memory usage Generally, memory for loaded data, indexes etc. is managed on a per-block basis, and capped by the block cache capacity. Note that this also applies to returned values: When you hold a `Slice`, it keeps the backing buffer alive (which may be a block). If you know that you are going to keep a value around for a long time, you may want to copy it out into a new `Vec`, `Box<[u8]>`, `Arc<[u8]>` or new `Slice` (using `Slice::new`). > [!NOTE] > It is recommended to configure the block cache capacity to be ~20-25% of the available memory - or more **if** the data set fits _fully_ into memory. Additionally, orthogonally to the block cache, each `Keyspace` has its own write buffer (["Memtable"](https://docs.rs/fjall/latest/fjall/struct.KeyspaceCreateOptions.html#method.max_memtable_size)) which is the unit of data flushed back into the "proper" index structure. ## Error handling Fjall returns an [error enum](https://docs.rs/fjall/latest/fjall/enum.Error.html), however these variants are mostly used for debugging and tracing purposes, so your application is not expected to handle specific errors. It's best to let the application crash and restart, which is the [safest way to recover from transient I/O errors](https://ramalagappan.github.io/pdfs/papers/cuttlefs.pdf). ## Transactional modes The backing store (`lsm-tree`) is a MVCC key-value store, allowing repeatable snapshot reads. However this isolation level can not do read-modify-write operations without the chance of lost updates. Also, `WriteBatch` does not allow reading the intermediary state back as you would expect from a proper transaction. For that reason, if you need transactional semantics, you need to use one of the transactional database implementation (`OptimisticTxDatabase` or `SingleWriterTxDatabase`). TL;DR: Fjall supports both transactional and non-transactional workloads. Chances are you want to use a transactional database, unless you know your workload does not need serializable transaction semantics. ### Single writer Opens a transactional database for single-writer (serialized) transactions. Single writer means only a single **write** transaction can run at a time. This is trivially serializable because it _literally_ serializes write transactions. ### Optimistic Opens a transactional database for multi-writer, serializable transactions. Conflict checking is done using optimistic concurrency control, meaning transactions can conflict and may have to be rerun. ## Feature flags ### lz4 Allows using `LZ4` compression, powered by [`lz4_flex`](https://github.com/PSeitz/lz4_flex). *Enabled by default.* ### bytes_1 Uses [`bytes`](https://github.com/tokio-rs/bytes) 1.x as the underlying `Slice` type. Otherwise, [`byteview`](https://github.com/fjall-rs/byteview) is used instead. *Disabled by default.* ## Stable disk format Future breaking changes will result in a major version bump and a migration path. For the underlying LSM-tree implementation, see: . ## Examples [See here](https://github.com/fjall-rs/fjall/tree/main/examples) for practical examples. ## Contributing How can you help? - [Ask a question](https://github.com/fjall-rs/fjall/discussions/new?category=q-a) - or join the Discord server: [https://discord.com/invite/HvYGp4NFFk](https://discord.com/invite/HvYGp4NFFk) - [Post benchmarks and things you created](https://github.com/fjall-rs/fjall/discussions/new?category=show-and-tell) - [Open a PR](https://github.com/fjall-rs/fjall/compare), - [See open issues to pick up here](https://github.com/search?q=org%3Afjall-rs+label%3A%22help+wanted%22+state%3Aopen+&type=issues) - [Open an issue](https://github.com/fjall-rs/fjall/issues/new) (bug report, weirdness) ## License All source code is licensed under MIT OR Apache-2.0. All contributions are to be licensed as MIT OR Apache-2.0.