HarperDB eliminates much of the complexity for IoT projects, and allows for distributed computing directly on the edge.
HarperDB is written in Node.js which requires significantly less RAM and CPU than a traditional database. In fact, at rest, HarperDB uses no resources as it is stateless. This is ideal for IoT Devices and IoT projects as it allows for battery conservation and the maximization of value of these micro-computing devices.
Interfacing with HarperDB is done through a simple to use REST API. HarperDB has a single end point with a consistent JSON object model that allows you to simply and explicitly interact with your data.
Built on a websocket framework leveraging socket.io, HarperDB has robust and performant clustering and replication capabilities. Replication is defined at a table level. HarperDB's clustering algorithm uses a democratized model allowing for each node to function as a master and all nodes to function as a slave. This avoids single points of failure eliminating the need for IoT Gateways, and is excellent in areas where internet may be unreliable. This makes HarperDB intuitive and easy to use for IoT applications.
“There are a few reasons why HarperDB stands out for IoT. It can be installed on any hardware due to its size, plus packs more functionality in less space. The solution allows us to manage thousands of IoT devices, combining structured and unstructured data, while leaving HarperDB to handle versioning and concurrency. We can also integrate our Visual Content Analysis data in a single database that can integrate itself with centralized databases for more complex A.I. processing."