MapD Core OverviewΒΆ

MapD Core is an in-memory, column store, relational database that works with native SQL. While designed from the ground up for GPU environments, it also runs on CPU hardware.

MapD Core is the foundational element of a larger data exploration platform that emphasizes speed at scale. By taking advantage of the parallel processing power of the hardware, MapD Core can query billions of rows in milliseconds. Furthermore, by using the graphics pipelines of GPUs, MapD Core can render graphics directly from the server.

MapD can operate as a standalone database using the command line or its administration console. MapD can also output to its native interface, MapD Immerse, or to third-party software such as Tableau, Qlik, or Birst, using a variety of connectors.

Within a single server, MapD Core can handle datasets 1.5-3TB in raw size in GPU RAM, or 10-15TB in CPU RAM. In distributed deployments, that number scales linearly.