About the Stats page

The Stats page allows you to monitor your real-time analytics and view your historical caching statistics for your services on the web interface.

Before you begin

Be sure you know how to access the web interface controls before learning about the details you'll encounter here.

About the Stats page

On the Stats page you can access the:

The Stats link appears automatically for logged in users with the appropriate access permissions.

Viewing the Real-time stats

The Real-time stats page allows you to separately monitor caching for each of your services in real time, as they operate on a second-by-second basis.

The data on this page may appear grayed out or blank to some users, with no information displayed in the controls, when a customer's service does not receive enough requests for Fastly to display meaningful information about it in real time. If you've just created your service, you might see a message saying there's nothing to see yet.

real-time cache activity for a service per second from all data centers

In addition to a menu allowing you to select the specific data center from which to view data (it defaults to data from all data centers), the top of the dashboard includes the following real-time cache activity:

  • Requests: the number of requests Fastly receives for your site per second.
  • Hits: the number of times requested data is found in cache and does not require making a fetch your origin server.
  • Misses: the number of times requested data is not found in cache and has to be requested from your origin server.
  • Hit Ratio: the percentage of content being accessed that is currently cached by Fastly. Also known as Fastly's definition of cache hit ratio. While there are many ways to calculate cache hit ratio, at Fastly this is defined as the proportion of cache hits (hits) to all cacheable content (hits + misses). If shielding is enabled, it can cause the hit ratio to be inaccurate.
  • Errors: the number of error responses per second that occur as Fastly receives requests for your site.
  • Hit Time: the average amount of time (in milliseconds) spent processing cache hits.
  • Miss Time: the average amount of time (in milliseconds) spent processing cache misses.

Below the real-time cache activity summary data, several graphs appear:

Individual dashboard graphs

The graphed cache activity includes:

  • Global POP Traffic: a heat map displaying global POP traffic through all POPs for your service.
  • Requests: a graph displaying the total number of requests received for your site by Fastly over time.
  • Errors: a graph displaying the number of error requests that occurred over time.
  • Hit Ratio: a graph displaying the percentage of content being accessed that is currently cached by Fastly over time.
  • Bandwidth: a graph displaying the bandwidth served from Fastly's servers to your website's visitors.
  • Image Optimizer: when enabled, a graph displaying the number of responses that came from the Fastly Image Optimizer service over time.
  • Image Optimizer (Videos): when enabled, a graph displaying the number of videos responses that came from the Fastly Image Optimizer.
  • Image Optimizer (Video Frames): when enabled, a graph displaying the number of video frames that came from the Fastly Image Optimizer. A video frame is an individual image in a sequence of images that make up a video.
  • Logs: a graph displaying the number of logs sent to your endpoints from Fastly.
  • Log Bandwidth: a graph displaying the total bandwidth size of the logs sent to your endpoints from Fastly.
  • Origin Latency: a histogram displaying the average amount of time to first byte (measured in milliseconds) on a cache miss or pass. High origin latency means that your backends are taking longer to process requests.

You may not see any traffic right away because of the following:

  • Not enough data is going to your site. If this is the case, visit the site yourself to trigger some traffic.
  • You've made a CNAME change. If this is the case, it could take from a few minutes to hours for the change to propagate your DNS servers. See how to edit your DNS record to point to Fastly for more information.

Once you start seeing real-time cache activity, you also can interact with your stats graphs.

Viewing the Historic stats

The Historic stats page provides a visual interface to our stats API for a selected Fastly service. You can also display historical metrics aggregated across all your Fastly services by clicking All services. The graphs display metrics derived from your site's statistical information. If you've just created your service, you might see a message saying there's nothing to see yet.

The displayed caching and performance metrics help you optimize your website's speed. These metrics include the following:

  • Hit Ratio metrics tell you how well you are caching content using Fastly. This metric represents the proportion of cache hits (hits) to all cacheable content (hits + misses). Increasing your hit ratio improves the overall performance benefit of using Fastly.
  • Cache Coverage metrics show how much of your site you are caching with Fastly. This metric represents the ratio of cacheable requests (i.e., non "pass" requests) to total requests. Improving your cache coverage by reducing passes can improve site performance and reduce load on your origin servers.
  • Caching Overview metrics compare Cache Hits, Cache Misses, Synthetic Responses (in VCL edge responses), and Passes (or requests that cannot be cached according to your configuration).

The traffic metrics analyze your website's traffic as it evolves over time. These metrics include the following:

  • Requests metrics show you the total number of requests over time that were received for your site by Fastly.
  • Bytes Transferred metrics show you the total number of bytes transferred by Fastly for your service.
  • Header & Body Bytes Transferred metrics show you the relative values of bytes transferred when serving the body portion of HTTP requests and the header portion of the requests.
  • Miss Latency metrics show the distribution of only the miss latency times for your origin.
  • Error Ratio metrics show you the ratio of error responses (4xx and 5XX status code errors) compared to the total number of requests for your site. This metric allows you to filter types of error responses and quickly identify error spikes at given times.
  • HTTP Info, Success, & Redirects metrics shows the number of HTTP Info (1XX), Success (2XX), and Redirect (3XX) statuses served for your site using Fastly.
  • Status 3XX Details metrics shows the breakdown between the number of HTTP Status 301s, 302s, 304s, and other 3XX requests.
  • HTTP Client and Server Errors metrics shows the number of HTTP Client Errors (4XX), and Server Errors (5XX) served for your site by Fastly.
  • HTTP versions metrics shows the number of requests using HTTP/1.1, HTTP/2, and HTTP/3 (QUIC) protocols. Note that HTTP/3 (QUIC) is only available as part of a limited availability release. To learn more, contact your account manager or email support@fastly.com for details.
  • TLS versions metrics shows the number of requests using TLS 1.0, TLS 1.1, TLS 1.2, and TLS 1.3 protocols.
  • Logs metrics show the number of logs sent to your endpoints from Fastly.
  • Log Bandwidth metrics show the total bandwidth size of the logs sent to your endpoints from Fastly.
  • When enabled, Image Optimization Requests metrics show you the number of responses that came from the Fastly Image Optimization service.

Once you start to see your caching and performance metrics, you also can interact with your stats graphs.

Working with stats graphs

You can interact with and control your Real-time and Historic stats graphs as follows.

Limiting data viewed to specific data centers

By default, Real-time graphs display data from all data centers. To view data from a single, specific data center, select it from the All datacenters menu.

Disabling smooth scrolling

The Real-time graphs update continuously. Leaving the graphs open for long periods of time, however, can occasionally lead to higher CPU utilization. To improve performance, you can deselect the Smooth scrolling checkbox. The graphing animations may not be as smooth when this checkbox is deselected.

Viewing the real-time stats timestamp indicator

Hovering the cursor over any part of a graph displays a timestamp indicator that updates itself as you move the mouse.

hover animation for graph timestamp

Hiding and displaying the mean value

A dashed line indicating the mean value of the graph's data appears on some graphs. To hide the mean line, deselect the Mean checkbox.

average graph line

Expanding and minimizing graph views

You can expand and minimize the view of some of the graphs using the quadruple arrow button in the right-hand corner of the graph to display an expanded view of the graph or special features it offers. Specifically:

  • the Global POP Traffic heat map displays a larger view of the graph as well as the traffic in each POP region, with continuously updating data on the POP's current requests per second, the request error ratio, and the bandwidth going through that POP.
  • the Requests, Errors, and Hit Ratio graphs expand to larger versions of themselves along with the already expanded versions of the Bandwidth and Origin Latency graphs.
  • the Origin Latency graph specifically includes a small gear icon in the upper right corner that allows you to change the interval limit displayed by the graph from the default 15 second interval to a shorter time frame.

Viewing service version activation

Service version activations appear as vertical lines on the Historic graphs. Hovering your cursor over any line displays the version's number and its activation timestamp.

stats hover for version activation indicator

Controlling the historic stats date displayed

You can control how you view the historic stats date ranges. For all displayed graphs, you can choose:

  • the exact local date and time range of the graphed data
  • how often to sample the data displayed
  • whether to view global data for the graphs or only data from a specific region
  • how to display the statistical values

Keep in mind, however, that data won't appear yet for time periods that haven't ended. Data aggregated per day is collected based on UTC days and each day's data becomes available around 2am the following day. Data aggregated by hour becomes available approximately 15 minutes after the end of each hour. Data aggregated by minute usually becomes available two minutes after the end of the minute, but can take up to 15 minutes. If your use case requires data closer to real-time, consider using the real-time stats instead.

Changing the stats displayed in historical graphs

You can change the statistics displayed in any historical graph to display an average, a 95th percentile, a minimum, a maximum, or a total. When set to average, the graph displays the average (mean) as a dashed line.

To change the statistics displayed by all historical graphs, use the Display menu in the upper right corner, as shown below.

controlling the types of displayed stats for all graphs via the Display menu

To change the statistics displayed for a individual graph, click the menu below the graph's name, as shown below.

Controlling the type of displayed stats for a single graph

You can also exclude certain data entirely from some graphs. For example, in the Status 3XX Details graph, you can click Show or Hide or the corresponding color ship next to the specific 3XX errors (301, 302, etc.) to show or hide those error types, respectively.

Requests overview graph with Hits hidden

What's next

Dig deeper into details about all areas of the web interface controls before you move on to using them to work with services.

Back to Top