Monitoring tool company Datadog has enhanced its Watchdog engine to provide insight into the health and performance of infrastructure components. The new addition promises to work with products like Redis or Amazon Web Services without requiring a special setup.
Watchdog was introduced in July 2018 to find anomalies like spikes in the APM metrics hits, error rate, and latency. This is done through machine learning approaches like anomaly and outlier detection, as well as forecasting algorithms.
The tool compiles notable findings such as the timeframe, affected resources, and duration of the irregularity in so-called stories, which link to detail pages that provide more context. Watchdog also hints at behaviour that might be related to help users assess the severity of a problem.
The now available infrastructure component adds functions to find patterns and trends in integrations concerning system metrics, Redis, PostgreSQL, NGINX, and AWS to the tool. Infrastructure that needs to be monitored can be added to Watchdog via its Monitors page.
According to the announcement, the service automatically detects the infra, so no special setup is required. Datadog customers are assured they will find everything needed to keep an eye on a certain source through a new category in the “New Monitor” section which also allows them to configure altering.
Another way of enabling a monitor is via Watchdog stories as they offer custom monitors. This is especially helpful if teams want to make sure they are notified should similar issues to the one reported in a certain story arise.