Datadog launches Updog.ai for real-time SaaS & AWS status monitoring

Datadog has released Updog.ai, a free public web page providing near real-time monitoring of the health of major AWS services and SaaS APIs.

Updog.ai is designed to give users independent, aggregated visibility into the health of popular platforms such as OpenAI, GitHub, Slack, Stripe, Zoom, ServiceNow, and Zendesk, as well as AWS services including S3, Lambda, and DynamoDB.

Unlike vendor-maintained status pages or third-party aggregators, Updog.ai uses anonymised telemetry data and AI models to generate timely status updates.

By correlating data from thousands of customer environments, the tool is able to highlight emerging performance issues and broad outages as they occur, rather than relying solely on announcements from service providers.

According to Datadog, this approach allows engineers to quickly determine whether an issue they are experiencing is isolated or part of a wider incident. It also enables faster identification of systemic outages. For example, Updog.ai detected a well-publicised Amazon DynamoDB degradation 32 minutes before AWS posted an update on its own status page.

Service coverage

Updog.ai provides a live dashboard that tracks the status of more than 30 SaaS providers and 13 AWS services. It also offers up to 90 days of historical incident data, making it easier for teams to track reliability trends and recurring disruptions, such as API issues affecting checkout processes.

This historical view helps organisations examine trends and makes it possible to inform architectural decisions, potentially improving fault tolerance and reducing the impact of third-party service outages.

Aggregated observability

The release of Updog.ai represents an extension of observability beyond individual customer environments. Datadog is aggregating and analysing telemetry data from across its entire customer base, creating a broader pool of insight into service health that would not be possible through a single organisation's data alone.

"Observability has traditionally been bound by the walls of individual systems, with teams focused on what they could measure within their own environments. Datadog is redefining that boundary by collecting and correlating telemetry data across the entire breadth of our products and customer base. With one of the world's largest and most diverse streams of telemetry data, we can apply AI models that identify patterns and risks that no single organization can see on its own. This represents a shift from simply helping customers manage their environments to creating shared intelligence."

This strategy aims to surface insights into systemic issues, allowing engineering teams to see error signals that may otherwise go undetected in isolation. By doing this, Datadog indicates it is supporting both its customers and the broader technology community with greater transparency regarding provider reliability.

Real-time status driven by AI

Updog.ai builds on Datadog's prior External Provider Status functionality by integrating advanced AI models to analyse aggregated and anonymised Application Performance Monitoring (APM) telemetry.

This process includes using a Bayesian model to infer abnormal error rates and cross-referencing signals across regions and customer environments to confirm if degradations are widespread.

Through this AI-driven approach, Datadog reports it can often detect issues significantly earlier than service owners update their status pages. The company notes this results in a service status signal more closely aligned with the real-world user experience.

Future developments

Datadog has outlined plans to expand Updog.ai's monitoring capabilities.

Upcoming features include monitoring GPU availability, allowing AI infrastructure teams to better plan workloads, as well as spot interruption monitoring aimed at helping infrastructure teams anticipate and mitigate spot interruptions.

In addition, the company plans to introduce cyber attack and vector monitoring, offering a perspective on global threats and attack vectors.

"Built on anonymized observability data and AI at internet scale, Updog.ai is a comprehensive public resource for real-time service transparency."

Updog.ai is free to use and does not require a Datadog account to access the live status dashboard of major providers.

The service aims to provide engineers and organisations with greater insight into third-party service reliability and performance, supporting both real-time troubleshooting and longer-term operational planning.