Components and terms for Amazon CloudWatch Internet Monitor

A monitor includes the resources for a single application that you want to view internet performance and availability measurements for, and that you want to get health event alerts about. When you create a monitor for an application, you add resources for the application to define the cities (locations) for Internet Monitor to monitor. Internet Monitor uses the traffic patterns from the application resources that you add so that it can publish internet performance and availability measurements specific to just the locations and ASNs (typically, internet service providers or ISPs) that communicate with your application. In other words, the resources that you add create a scope of the city-networks that you want Internet Monitor to monitor and that you want it to publish measurements for.

A resource that you add to a monitor is a "monitored resource" in Internet Monitor. That is:

In Internet Monitor, an ASN typically refers to an internet service provider (ISP), such as Verizon or Comcast. An ASN is a network provider that a client uses to access your internet application. An Autonomous System (AS) is a set of internet routable internet protocol (IP) prefixes that belong to a network or a collection of networks that are all managed, controlled, and supervised by one organization.

A city-network is the location (such as a city) that clients access your application resources from and the ASN, typically an internet service provider (ISP), that clients access the resources through. To help control your bill, you can set a limit for the maximum number of city-networks for Internet Monitor to monitor for each monitor. You pay only for the actual number of city-networks that you monitor, up to the maximum number. For more information, see Choosing a city-network maximum limit.

Internet Monitor also publishes internet measurements to log files in CloudWatch Logs every five minutes for the top 500 city-networks for your monitored application traffic.

These measurements quantify the performance score, availability score, bytes transferred (bytes in and bytes out), and round-trip time for your application's city-networks. These are measurements for the city-networks specific to your VPCs, Network Load Balancers, CloudFront distributions, or WorkSpaces directories. Optionally, you can choose to publish internet measurements and events for all monitored city-networks (up to the 500,000 city-networks service limit) to an Amazon S3 bucket.

Internet Monitor generates aggregated metrics for CloudWatch metrics, for global traffic to your application and global traffic to each AWS Region. For more information, see View Internet Monitor metrics or set alarms in CloudWatch Metrics.

Internet Monitor creates a health event to alert you to a specific problem that affects your application. Internet Monitor detects internet issues, such as increased network latency, across the world. It then uses its historical internet measurements from across the AWS global infrastructure footprint to calculate the impact of current issues on your application, and creates health events. Internet Monitor, by default, creates health events based on both overall impact and local impact thresholds. To learn more about health events, see When Internet Monitor creates and resolves health events.

The default health event threshold, for both performance scores and availability scores, is 95%. If you like, you can specify your own custom thresholds for when Internet Monitor creates health events. For more information about configuring thresholds, see Change health event thresholds.

Each health event includes information about the impacted city-networks. You can view health events in the CloudWatch console, or by using an AWS SDK or AWS CLI with Internet Monitor API actions. Internet Monitor also sends Amazon EventBridge notifications for health events. For more information, see When Internet Monitor creates and resolves health events.

Internet Monitor displays information about recent global health events, called internet events, on an internet weather map that is available to all AWS customers. You don't need to create a monitor in Internet Monitor to view the internet weather map. Unlike health events, internet events are not specific to individual customers or their application traffic. For more information, see Global internet weather map in Amazon CloudWatch Internet Monitor.

Internet Monitor creates health events based on both overall thresholds and local thresholds. You can change the default thresholds and configure other options, such as turning off local thresholds. For more information about configuring thresholds, see Change health event thresholds.

By analyzing the data that AWS collects, Internet Monitor can detect when the performance and availability for your application has dropped, compared to estimated baselines that Internet Monitor calculates. To make it easier to see those drops, Internet Monitor reports the information to you as scores. A performance score represents the estimated percentage of traffic that is not seeing a performance drop. Similarly, an availability score represents the estimated percentage of traffic that is not seeing a availability drop. For more information, see How AWS calculates performance and availability scores.

Bytes transferred is the total number of bytes of ingress and egress traffic between an application in AWS and the city-network (that is, the location and the ASN, typically the internet service provider) where clients access an application. Monitored bytes transferred is a similar metric, but includes only bytes for monitored traffic.

Round-trip time (RTT) is how long it takes for a request from a client user to return a response to the user. When RTT is aggregated across client locations (cities or other geographies), the value is weighted by how much of your application traffic is driven by each client location.