Application Metrics

Application-level metrics help developers and operations staff identify, investigate, and diagnose issues. Metrics provides key attributes of app health including response time, throughput, errors, dyno load, and memory, unified on a single time axis so you can easily see correlations. Accessible via the Heroku Dashboard, Application Metrics are available on all paid dynos. Learn more >>

Threshold Alerting

Two of the most important indicators of app health are: app responsiveness, measured by the max response time for 95% of requests, and the number of failed requests. Threshold Alerting lets you specify limits on web dyno 95th percentile response time and the percentage of failed requests, above which an alert will be triggered. Email, PagerDuty, and dashboard notifications are supported. Threshold Alerting is available to apps running on Professional or Private dynos. Learn more >>

Autoscaling

Autoscaling horizontally scales your app’s web dynos to meet your specified 95th percentile response time threshold, based on your existing throughput. With autoscaling there’s no need to anticipate your traffic spikes. Autoscaling is included for free on Performance and Private dynos. Learn more >>

Consolidated Logs

The View Logs feature allows you to tail and view the logs for your application within the Dashboard, providing more in-depth details about events and errors. Log tails are also accessible via the command line interface (CLI). You can also drain your logs to a drain service provided by one of the many logging add-on providers in the Elements marketplace for long-term storage, search, filtering and troubleshooting. Learn more >>

Discover Heroku OpEx
Visibility Into App Health

With up to 7-day metrics retention, Application Metrics helps you view trends and quickly identify anomalies in application health.

Proactive Notifications

We watch your app, so you don’t have to. Heroku ends notifications if the response time or error rate for your web dynos exceeds your criteria.

Effortless Scaling

No need to keep an eye on your traffic spikes anymore, we automatically adjust the number of web dynos to meet your specified latency threshold.

Performance Tuning Guidance

With features like language-specific Application Guidance, we help you identify and correct application problems.

No Additional Cost

Metrics are available for Hobby dynos and above at no additional cost. Advanced features are unlocked as you move to higher level dynos.

  • We trust Heroku’s Autoscaling to handle traffic spikes, so we don’t have to worry about monitoring our apps and managing scale.

    Alpine Lab Michael Baudino Founder, Alpine Lab

  • The Heroku platform suits our remote team’s work style. Anyone on the team can easily manage or monitor our apps asynchronously, and automated features mean that nothing gets hung up while some team members are offline.

    Calorie Cloud Troy Hickerson Co-founder, Calorie Cloud Read customer story >>

  • Anything we can do on Heroku, we will do, because it simplifies operations so much.

    LendUp Jake Rosenberg CTO, LendUp Read customer story >>

  • Running our client-facing app on Heroku allows us to seamlessly scale our operations to meet the needs of a growing client base.

    RP/6 Kylee Durant COO, RP/6 Read customer story >>

  • We’re big fans of the new Heroku Dashboard with metrics that let us easily monitor and fine-tune the performance of our apps. It saves us a lot of time.

    Product Hunt Mike Coutermarsh Software Engineer, Product Hunt Read customer story >>
Documentation
Podcast

A streamlined approach to application operations, and how Heroku built autoscaling

Listen to our podcast with Software Engineering Daily from February 28th, 2017.

In this podcast Andrew Gwozdziewycz, Heroku operational experience engineer, talks about the importance of application health, Heroku's metrics platform architecture and how it laid the foundation for autoscaling.