Flexible
77 |78 | Start saving your data today, then adapt your queries as your application evolves, without expensive schema migrations. 79 | Query your data using familiar JavaScript (or your language of choice). 80 | CouchDB's incremental Map Reduce views are optimized for real-time workloads, not overnight batch jobs. 81 |
82 |Web scale
85 |86 | CouchDB scales from a smartphone to a data-center, and beyond. 87 | The Erlang and REST-based implementation presents the same, simple API 88 | whether running on a small local instance, 89 | an ad-hoc network of collaborators, 90 | or a multi-terabyte eventually-consistent cluster. 91 |
92 |Reliable
95 |96 | CouchDB uses append-only storage for extreme durability. It never touches any bytes already written to disk, so even truncating a CouchDB database file yields a consistent snapshot of an older version of the database. 97 | Writes are contiguous, giving CouchDB predictable performance even under heavily concurrent write load. The MVCC data model is ideal for concurrent operations that require consistency. 98 |
99 |Ground computing
102 |103 | CouchDB solves the data-island problem, so you can write offline capable applications. 104 | Peer-based replication allows for flexible deployments, data-sharing between organizations, and no-hassle backups. 105 | CouchDB on the client offers better latency and reliability than traditional three-tier architectures. Ground computing is the future of the web. 106 |
107 |Trusted
110 |111 | CouchDB is deployed by the BBC, Meebo, Cloudant and Ubuntu One. 112 | It ships as a core feature of Ubuntu Linux. 113 | There are access libraries available in nearly every language. 114 |
115 |