Datadog has helped Konotor, a mobile-first user-engagement platform, solve its problem of scaling its PostgreSQL database by detecting issues before they affect users and correlating different events and metrics to make sense of what's happening. The company discovered that it had exceeded the IOPS (input/output operations per second) limit on its SSD-backed AWS Elastic Block Store, causing slow writes. After investigation, they found that the issue was due to incorrect calibration of IOPS allocated to the disks. Datadog provided valuable insights and tools, such as monitoring CPU time spent on I/O waits, disk latencies, and correlating data, which helped Konotor implement solutions like striping multiple disks, using Provisioned IOPS (SSD), ephemeral storage, moving frequently used tables/indexes to another disk, reducing I/O writes, detection and prevention measures. Datadog's lightweight agents and plugins also made it easy for Konotor to monitor its infrastructure and identify potential problems before they affect users.