A platform for high-performance distributed tool and library development.


High Performance
C++ API and no JVM overhead means fast data processing.

Unique in-place data allocation means no serialization/deserialization cost.

Run computations in a laptop or in a cluster with tens or hundreds of machines (easy to deploy).

Easy to Use
Store and manipulate complex objects in the cloud using simple object-oriented/functional API.

 * For details see PlinyCompute paper on arXiv