Voldemort is an open source distributed database based on Amazon Dynamo. Voldemort uses in-memory caching to eliminate a separate caching tier. It has a storage layer that is possible to emulate. Voldemort reads and writes scale horizontally. The API decides data replication and placement and accommodates a wide range of application-specific strategies. The Voldemort distributed data store supports pluggable placement strategies for distribution across data centers. Data is automatically replicated across servers. Data is partitioned meaning a single server contains only a portion of the total data. Each data node is independent to avoid central point of failure. Pluggable serialization allows rich keys and values including lists and tuples with named fields, as well as the integration with common serialization frameworks such as Avro, Java Serialization, Protocol Buffers, and Thrift. This serialization type uses a JSON data model but a more compact byte format, and also checks data against an expected schema for type correctness. Data items are versioned, which maximizes data integrity. Server failures are handled transparently. The storage layer is completely mockable so development and unit testing can be done against a throw-away in-memory storage system without needing a real cluster (or even a real storage system) for simple testing. Voldemort supports hashable semantics; this means that a single value can be modified and that the search is done by primary key. This makes distribution across machines particularly easy since everything can be split by the primary key. Voldemort has proven to be a highly available and partition safe while maintaining high throughput.
Voldemort
