Startup HarperDB Inc. is emerging from stealth mode today with a faster database specifically for software developers creating applications for the “internet of things.”

Combining traditional SQL for structured data and NoSQL for other kinds of data such as documents or photos, HarperDB is a hybrid transactional/analytical processing database. Gartner Inc. defines these HTAP databases as an emerging application architecture that “breaks the wall” between transaction processing and analytics, enabling more informed and real-time decision making.

One of the main advantages of HTAP is that it allows advanced analytics to be run in real time on “in flight” transaction data, providing an architecture that empowers users to respond more effectively to transient opportunities.

HTAP databases aren’t new. Well-known examples include Aerospike, MongoDB and DataStax. But although these alternatives are extremely good at ingesting high-scale data, they often struggle with analytics because of their lack of SQL capabilities, Stephen Goldberg, HarperDB’s chief executive, told SiliconANGLE.

The problem with these databases is that when users attempt to perform SQL operations, they first need need to transform that data either on disk or in memory into something resembling a column or row. Doing so consumes significant resources and also renders real-time data analytics impossible, as it takes time to prepare that data.

That’s where HarperDB has an advantage, Goldberg said. HarperDB comes with a little secret sauce in the shape of its patent-pending data storage algorithm, which can ingest both structured and unstructured data and transform it into a single-model data store, with no configuration required. This eliminates the need to have and manage multiple databases, enabling developers to build applications that can interact with their data in real-time.

HarperDB touts its database for a number of use cases, but it’s primarily focused on IoT deployments, Goldberg said. Thanks to its small footprint, HarperDB can easily be installed onto low-powered micro-computing devices at the network edge that predominate in IoT systems. By doing so, HarperDB enables distributed computing at the network edge that can easily scale out.

“HarperDB is not a platform but rather a stand-alone database with a very small footprint,” Goldberg explained. “We are focusing on IoT right now primarily because HarperDB as an HTAP database can be installed directly on an edge device where resources like memory and CPU are limited.”

The company also reckons its database is an ideal foundation for application development. Not least, HarperDB said, its database can be used by larger enterprises as a schemaless data warehouse that allows developers to perform structured queries on unstructured data in real time.

HarperDB is offering two versions of its database, a free community edition and an enterprise version for larger-scale projects.

Image: HarperDB