A distributed, on-disk key-value store that scales RocksDB with seamless cluster resizing and eventual consistency.
Vasto is a distributed key-value store that scales embedded RocksDB into a distributed system with sharding, replication, and dynamic cluster management. It solves the problem of managing distributed storage with simple commands, offering high availability and eventual consistency without service interruptions during scaling.
Developers and teams needing a scalable, self-hosted key-value store for applications requiring persistent storage with dynamic cluster resizing and high availability.
Developers choose Vasto for its simplicity in managing distributed storage, seamless scaling operations, and efficient architecture that avoids complex consensus protocols while providing low latency and eventual consistency.
A distributed key-value store. On Disk. Able to grow or shrink without service interruption.
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Allows growing or shrinking the cluster without service interruption, using efficient data migration via SSTable writes and jumping consistent hashing for minimal data movement.
Managed via simple commands for creating, deleting, and resizing keyspaces, abstracting away the complexity of partitioning and replication as highlighted in the philosophy.
Uses a single master without gossip protocols or consensus algorithms, enabling fast setup and failure detection with soft state recovery on restart.
Supports reads and writes on all replicas with changes propagated within milliseconds, and conflict resolution based on event timestamps for distributed environments.
The gateway for non-Go languages is explicitly stated as 'not ready yet' in the README, restricting direct use to Go applications and hindering adoption in polyglot systems.
Relies on a single master for topology changes; while it holds soft state and can restart, it remains a potential bottleneck and failure point in high-availability scenarios.
Conflict resolution depends on accurate event timestamps, which can be problematic with clock skew, and may not suit use cases requiring immediate data consistency guarantees.