With Datadog’s Edgecast integration, you can gain visibility into your CDN operations, monitor the impact of cache configuration changes in pre-production, and deploy with confidence. In this post, we’ll cover how to visualize Edgecast network performance metrics with an out-of-the-box (OOTB) dashboard and get alerted on key performance issues, such as a low cache hit rate ratio or an anomalous increase in traffic and errors.
The Edgecast integration includes a dashboard that allows you to visualize key performance and traffic metrics, such as your network’s cache hit ratio and bandwidth. Datadog automatically tags Edgecast metrics by origin, making it easy to analyze aggregate data and then home in on a specific host. To see a full list of metrics collected by this integration, refer to our documentation.
Graphs such as “Weekly Hit Ratio Change” give you insight into long-term shifts in your network. For example, sliding trends in your cache hit ratio may indicate larger, looming application issues. When a cache miss occurs, the request is passed along to the underlying data store—at scale, this can increase server load and latency when users attempt to load a webpage. Enabling features such as partial cache sharing can improve caching performance by delivering content in parts, allowing users to view content before it’s fully cached. Using the Edgecast dashboard, you can monitor how these policy changes affect your cache hit rate over time.
To visualize your Edgecast metrics in context with the rest of your infrastructure, you can clone the dashboard and customize it with metrics from any of the 600+ services Datadog integrates with. For example, you can view your Edgecast network metrics alongside web server data from Apache or NGINX, as shown above. By cross-examining your network traffic with how your web server handles requests, you can observe how your server handles spikes in bandwidth. This can help you determine whether you need to scale up the number of worker processes your server deploys.
To quickly detect and resolve issues before they impact customers, you can alert on key thresholds for metrics like cache hit ratio and average bandwidth. You can also configure anomaly monitors to notify you whenever a metric exhibits abnormal behavior that deviates from its historical baseline. Using anomaly monitors, you can intercept potential issues that would otherwise go undetected.
For example, by configuring an anomaly monitor for the number of requests that return 4xx or 5xx HTTP status codes, you can find out if new deployments contain broken links or if your users are encountering an abnormally high number of server-side errors. If you get notified, you can pivot to your Edgecast dashboard to investigate and identify the scope of impact.
Datadog’s Edgecast integration provides deep visibility into your CDN, and enables you to monitor its performance alongside your web servers and the rest of your infrastructure. By configuring alerts, you can also rapidly respond to abnormal spikes in traffic or errors to ensure a smooth customer experience.