API Slowdown - At 05:50 CET a number of queries came in with high data-volume filters. This caused CPU usage to spike as new workers was forced to activate on each host while the active workers where waiting for data, further slowing down response times.
At 05:55 CET the overall response times was back to normal, only a few uncached requests came in over the next 10 minutes.
The issue is fully resolved and our API has full performance and stability back.