Introducing Datadog Synthetics
Datadog Synthetics is now available. Start a free Datadog trial here.
Datadog is pleased to announce the availability of Synthetics, a whole new layer of visibility on the Datadog platform. By monitoring your applications and API endpoints via simulated user requests, Synthetics helps you ensure uptime, identify regional issues, track application performance, and manage your SLAs and SLOs. By unifying Synthetics with your metrics, traces, and logs, Datadog allows you to observe how all your systems are performing as experienced by your users.
API tests in Datadog Synthetics generate active HTTP requests (GET, POST, PUT, etc.) to your web properties or application endpoints at periodic, configurable intervals from multiple locations around the world. These checks verify not only that your applications are up and responding to requests, but that any conditions you define are met, such as response time, HTTP response code, and header or body contents.
This end-to-end visibility is useful for validating that end users can perform critical business transactions, such as visiting your homepage, signing up, and logging in. Misconfigured DNS, CDN, or web server settings can produce situations where your application looks fine internally but is actually inaccessible to some segment of your customers.
Synthetics was designed from the ground up to be integrated tightly to the rest of the Datadog platform. By monitoring the results of your synthetic tests in Datadog dashboards alongside application performance metrics and infrastructure metrics, you can get the full picture of user-facing performance in one place. That means that you can assess the scope of a network issue or quickly identify the start of an incident without needing to jump between tools. And when Synthetics alerts you to a user-facing outage or performance degradation, you have all the data you need to investigate and troubleshoot the issue without switching contexts.
Automatic, continuous URL testing
APM is a powerful tool for identifying issues and getting to root causes, but it relies on real user traffic to generate monitoring data. With the addition of Synthetics, Datadog can carry out periodic checks of your web applications and endpoints, whether or not actual users are currently accessing those resources. The continuous and automated nature of synthetic tests is especially valuable for identifying issues during off-hours and for monitoring resources that experience relatively little user traffic, such as internal business applications. Synthetic tests provide the availability and performance data you need to manage and track your SLAs and SLOs. Additionally, synthetic monitoring can help you validate the performance and availability of new endpoints or web pages proactively, before they are released to real users.
Minimize alert fatigue
Once you enable Synthetics, you can use all of Datadog’s monitoring and alerting functionality to notify your teams of any issues that might impact your users, while minimizing false positives. You can set up finely tuned alerts based on your data from Synthetics, including composite monitors to add conditional logic to your alerts. And our built-in integrations with collaboration tools like Slack and PagerDuty ensure that you can route notifications and alerts to the right teams.
A whole new layer of visibility
By providing a simulated user’s view of your application, Datadog Synthetics provides high-level performance insights that can quickly jump-start your investigation. An alert from Synthetics can point you to the exact application, endpoint, or region that is experiencing issues. You can then drill down and identify the root cause for any single synthetic request with an out-of-the-box integration with Datadog APM, and also view relevant host-level metrics and logs with a single click. And because Synthetics is focused on identifying symptoms, not causes, of performance problems, alerts built on synthetic tests can catch a wide array of issues, without the need to manually configure checks for every possible failure mode.