STRENGTHS
- Very fast reads after startup
- Straightforward single-file portability
- Efficient sequential write path
- Simple environment reconstruction
STORAGE / PROFESSIONAL
Single-file replay
Blindingly fast in-memory operation with efficient, fully persisted single-file storage.
01 / DESIGN
How Single-file replay stores data.
Single-file replay appends account, file and item mutations to one portable file. Startup replays that history into in-memory account, file, item and index structures; later reads operate primarily from memory, and compaction removes superseded history.
AVAILABILITYHigh-performance Professional storage option.
- 01Mutation
- 02Append record
- 03Single durable file
- 04Startup replay
- 05In-memory indexes
02 / TRADE-OFFS
Choose the engine for the workload.
Storage engines balance memory, I/O, recovery, concurrency and operational complexity. These profiles show where each engine fits.
WEAKNESSES
- Live indexes and values must fit in memory
- Startup cost grows with the replay history
- A very large working set increases heap pressure
- Compaction needs temporary space and coordinated rewriting
03 / ABILITIES
What it provides.
Single-file replay is built for production systems where speed is the number-one requirement and the working set fits in memory. Data remains fully and reliably persisted to disk, with an efficient persistence path designed to stay out of the workload's way.
Append-only mutation records
Versioned file header and structural limits
Startup replay and automatic compaction
Stored secondary-index data
BEST FIT
- Speed-critical production systems
- Workloads that fit in memory
- Low-latency application access
- Fast environment reconstruction
CORE CAPABILITIES
- In-memory operating speed
- Reliable full-disk persistence
- Efficient single-file storage
- ExecutionContext integration
CONFIGURATION
MVDB_SINGLE_FILE_DATA_ROOTMVDB_SINGLE_FILE_NAMEMVDB_SINGLE_FILE_MAX_ACCOUNTSMVDB_SINGLE_FILE_MAX_FILES_PER_ACCOUNT