Relational database management systems have been the mainstay of IT departments since they were first commercially introduced by Oracle in the 1970s. But over the past decade, a new generation of database categories have taken over as the top choice for enterprise IT projects.

In the past few years, developers have started to reach a consensus about which of these “specialty database categories” can best handle specific use cases where relational databases haven’t kept up. According to DB-Engines, the database management categories with the fastest user growth in the past two years are Time Series databases, Graph databases, and Key-Value stores. During that time, relational database usage has stayed the same or declined slightly.

Organizations now track, measure and analyze metrics from a wide range of sources that provide new data every second. According to IDC, the amount of data being produced worldwide is expected to grow nearly fivefold by 2025 to 175 zettabytes per year, driven by the proliferation of IoT sensors, serverless infrastructure, containerization and microservices.

Most of this is time-stamped data generated at high frequency and in great volume that requires rapid ingestion and real-time querying to extract maximum value. This means that the future of tech innovation will require real-time system observability, with granular insights to make more precise decisions to optimize operations and improve customer experience.

Paul Dix, creator of InfluxDB (open-source time-series database) and the founder and CTO of InfluxData joins me on Tech Talks Daily. With over two decades of experience helping companies like Microsoft, Google, McAfee, Thomson Reuters, and Air Force Space Command build software, Paul has unique insights to share on how and why time-series data is going to be the biggest hurdle and opportunity for tech organization in the future.