We’re excited to announce a major enhancement to our platform: trace-based alerts and metrics that better reveal the user experience and business performance. A trace tells the story of a transaction as it moves through multiple components of the system. A trace can show you the transaction components, and metrics tell you how the components are performing. 

Trace-based metrics are closer to application performance than AWS or K8s metrics, which focus on infrastructure-oriented data, such as CPU and memory usage. Alerting is one of the most important features in observability, enabling users to monitor application performance, know when issues are occurring, and troubleshoot/resolve those issues quickly. In addition, effective alerting enables teams to stay ahead of any issues that impact the user experience and business performance. 

Richer Data and Context for Monitoring & Alerting

Epsagon’s distributed tracing, trace-based metrics, and trace-based alerts enable you to quickly understand the user’s experience with an application, as well as the business implications of that user experience. Our approach is not only automated in its collection of trace data, but the breadth and depth of collected data enrich Epsagon’s monitoring and alerting capabilities, as well as enabling troubleshooting. Epsagon’s customers also have the ability to add tags to traces and collect specific information unique to customer events.

Using Epsagon’s trace-based metrics, users can monitor technical issues that can result in a user or business impact. Here are a few important use cases:

  • Latency issues when reading data from a database for a specific service or application
    • Users can create a trace-based alert for latency within a business billing application.
  • High error rate when servicing requests for paying customers 
    • Users can create an alert based on traces, which can correlate error rates and paying customers. This cannot be done with lambda functions or AWS metrics.
  • Low number of products placed in carts by users  
    • Users can create an alert with Epsagon traces, which can correlate metrics on the low number of products and the user id.

With Epsagon’s trace-based metrics, users can not only configure alerts but also receive proactive notifications via a preferred communications channel such as Slack, PagerDuty, or Microsoft teams when applications or services are underperforming. 

In our “Create Alert” view, users can alert on Lambda events and metrics, Kubernetes, and now also traces.

Creating traces based alerts with Epsagon

Epsagon’s ‘Create New Alert’ View

More Monitoring Enhancements from Epsagon

Trace-based metrics and alerts are part of Epsagon’s ongoing series of announcements that make monitoring more robust and easy-to-use.  

With the new cluster monitoring views for Azure Kubernetes Service (AKS), Amazon ECS, and Amazon Elastic Kubernetes Service (EKS), users and teams can understand: 

  • Cluster-level metrics over time for a specific cluster 
  • Nodes, pods, containers, and deployments associated with the specific cluster 
Epsagon's Kubernetes Dashboard with Node View

Epsagon’s Kubernetes Dashboard with Node View

We’ve also added to our platform:

  • High-level infrastructure monitoring dashboards with compute service metrics 
  • Custom monitoring dashboards to understand the overall health of an application
  • Expanded monitoring with auto-alerting and alerting on more metrics (including trace, custom, and AWS resource metrics) 

Try out trace-based metrics and alerts, along with our other recent monitoring enhancements, with a free 14-day trial.