FoundationDB develops an industrial‑strength, distributed transactional database whose data model is similar to Google’s F1. Its transaction core—branded “YesACID”—provides ACID guarantees and enables layered interfaces that can behave like MongoDB, SQL, or a key‑value store. The system is designed to be scalable and fault tolerant, able to replicate synchronously across datacenters with strong consistency and no single point of failure. The company acknowledges tradeoffs in complexity and higher commit latency from cross‑datacenter replication, which it says can be mitigated by bandwidth and parallel query throughput. FoundationDB is beginning to open‑source all of its layers for general use. The startup recently raised funding in a Series A to continue development following a significant seed round and investments from the founders and other early backers. FoundationDB is a modern key-value database that combines NoSQL-style scalability with ACID transactions, an approach the company calls “YesACID.” It is designed as an industrial-strength, fault-tolerant foundation for different data models that can be layered on top. The service is built to scale across distributed infrastructure without a single point of failure and to support transactions alongside horizontal scalability. Primary use cases cited include web applications facing scaling challenges, with the company referencing early scaling problems at Twitter. FoundationDB is available in alpha, plans to enter beta by the end of the year, and intends to offer a downloadable product plus an open ecosystem for libraries and layers. The founders have prior exits: they sold Visual Sciences in 2006 for about $60 million and that business was later acquired by Omniture in 2008 for $394 million. The company says its storage technology could be compatible with existing NoSQL APIs such as MongoDB.

Total raised

$17M

Funding rounds

2

Latest round

Series A

Latest activity

Nov 2013