Amazon CloudWatch Internet Monitor is a service that provides visibility into how internet issues impact the performance and availability of your applications hosted on Amazon Web Service (AWS). New Relic is a comprehensive observability platform that unifies monitoring, logging, and tracing across your entire technology stack. By offering real-time insights and analytics, New Relic helps businesses ensure optimal performance, reliability, and user experience. 

Today, New Relic is announcing its integration with Amazon CloudWatch Internet Monitor, providing customers with a unified, comprehensive solution for monitoring all applications and infrastructure telemetry. This integration simplifies the complexities of managing modern applications and infrastructure, driving improved business outcomes. By delivering real-time insights into internet performance and availability, businesses can swiftly identify and resolve issues that affect the end-user experience.

Set up

We have two options for reporting metrics data. The recommended path is setting up our Amazon CloudWatch Metric Streams integration. Alternatively, you can use our older Amazon integration that relies on polling. 

Logs data can be reported to New Relic using our Log ingestion Lambda. Be sure to add triggers to the Lambda and select the log group corresponding to internet-monitor at country, city, subdivision, and metro level. 

Observe

New Relic’s powerful dashboards and alerting systems enable teams to quickly identify and resolve issues, reducing downtime and improving operational efficiency. By tracking IP prefixes, autonomous system numbers (ASNs), latency, availability, health events, geographical data, and traffic patterns, it offers real-time insights into the health of internet services. 

Amazon CloudWatch Internet monitor provides the following telemetry and data points. 

  • Health event: When Internet Monitor detects significant performance degradation in your traffic, it creates a health event. Each health event includes information about the impacted client locations and network providers (ISPs).
  • Performance and availability scores (health scores): A statistical estimate of the percentage of traffic to your application that is not experiencing a performance or availability drop, respectively. These scores are also available as CloudWatch metrics.
    • An availability score represents the estimated percentage of traffic that is not seeing an availability drop. Internet Monitor estimates the percentage of traffic experiencing a drop from the total traffic seen and availability metrics measurements. For example, an availability score of 99% for an end user and service location pair is equivalent to 1% of the traffic experiencing an availability drop for that pair.
    • A performance score represents the percentage of traffic that is not seeing a performance drop. For example, a performance score of 99% for an end user and service location pair is equivalent to 1% of the traffic experiencing a performance drop for that pair.
  • Bytes transferred and monitored bytes transferred: 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: 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.

Alerts such as lower performance score and availability score can be integrated with tools such as slack, PagerDuty, etc.