MET Norway ensures free and real-time access to weather data with OCI

Imagine if thousands of entities and millions of individuals relied on your organization’s information to protect life and property. But suddenly, your ability to deliver that information was jeopardized by a critical equipment failure, and you narrowly avoided catastrophe by repairing that equipment with parts salvaged from a junk yard.

The Norwegian Meteorological Institute (MET) experienced such a scare in 2021. Established in 1866, MET collects and stores weather and climate data, conducts related research, and freely shares it with the world. For example, search and rescue agencies may analyze MET’s data to account for extreme weather when planning a rescue operation.

For over 20 years, MET’s knowledgebase was in an Oracle Database on an on-premises server. Users accessed that information via a bespoke application programming interface (API) that had become heavily modified and difficult to maintain. With the server also at end of life, system performance and operational risks became growing concerns as global demand for weather and climate data grew.

And then, during the 2021 winter holiday season, one of the server database disks crashed. Luckily, the downtime was minimal because IT staff repaired the server with cards found in similar hardware that had been retired. But for MET’s leadership and IT staff, it was a wake-up call.

Immediately following the server database crisis, MET contacted Oracle, its longtime partner, to get help with a short-term fix. The situation was complicated by the fact that IT staffers were in the process of testing a new open source NoSQL Apache Cassandra database that still had some stability issues.

Oracle’s BYOL cloud pricing model as well as pay-as-you-go pricing for consumption used was an opportunity to lower IT support costs. The scalability on-demand for peak reporting periods and the ability to handle data volumes and different types of data to account for seasonality were also key factors in MET’s decision.

Over the course of three months, MET’s IT staff worked with Oracle Consulting to spin up Oracle Autonomous Data Warehouse on Oracle Cloud Infrastructure (OCI), which eliminated complex database administration activities such as provisioning, patching, and tuning by applying machine learning automations. The teams worked through many challenges but ultimately established a reliable and secure environment.

Oracle Autonomous Data Warehouse on OCI is now the primary home for MET’s invaluable knowledgebase. IT staffers set up a proxy server to handle communication between the API and the data. Overall system speed increased by 10X with real-time query responses in milliseconds for a better user experience. Previously, with manual database tuning and on-premises infrastructure, response times could be as high as 5 to 10 seconds. In addition, eliminating the software licenses associated with the legacy database lowered IT support costs.

MET’s database administrator is now free to focus more strategically and creatively on the quality of weather and climate observations, not on database and server maintenance. Most importantly, self-patching in the autonomous cloud database solution eliminates downtime, calming any concerns of potential system failures. MET’s employees and users now have access to Norway’s premier climate and weather data 24/7.

Oracle Consulting used Oracle Zero Downtime Migration to manage the 10.2 Database migration, with both load and query going to the cloud.