What semantic monitoring can and can't do for microservices Overcome these challenges to detect anomalies in IT monitoring

Compare CloudWatch vs. Datadog and New Relic for AWS monitoring

Amazon CloudWatch continues to be a go-to monitoring tool for AWS resources, but hybrid cloud users can benefit from third-party options, such as Datadog and New Relic.

AWS continually evolves and expands its cloud platform. While these efforts potentially lead to feature bloat, and a services portfolio that's complex to navigate, they have also strengthened certain AWS offerings -- such as the vendor's monitoring and logging tools.

AWS introduced CloudWatch a decade ago, when there were still gaps in its cloud monitoring portfolio. The provider offered functionality to monitor cloud infrastructure, but lacked native tools to support composite applications comprised of cloud and on-premises components.

AWS users instead relied on third-party infrastructure and application performance monitoring (APM) software, such as New Relic, Datadog and others that adapted their software to handle various Amazon cloud services and workloads. However, AWS has continued to improve CloudWatch and its other native monitoring and management services. The result is a buyer's market for tools that monitor AWS resources and manage cloud-hosted apps.

Here, we'll evaluate CloudWatch vs. Datadog and New Relic.

CloudWatch is the baseline

Amazon CloudWatch is the all-purpose monitoring service for AWS. The tool provides a unified view of AWS resources and applications, including those that run on premises in Outposts.

CloudWatch features fall into five broad categories:

  • Data collection. CloudWatch aggregates logs and collects built-in AWS resource metrics, such as EC2 CPU utilization, API Gateway latency or container infrastructure data.
  • Monitoring. CloudWatch provides a dashboard that centralizes all operational data. This function also includes alarms for abnormal readings, log and metrics correlation, and Application Insights for SQL Server and .NET workloads. Custom dashboards and alarms help IT admins correlate data and events across resources and flag anomalies.
  • Action. CloudWatch Events or metrics can automatically trigger management tasks, such as auto scaling an EC2 instance pool, or custom actions carried out by command-line interface scripts, Lambda functions or SNS topic notifications.
  • Analysis. CloudWatch can graphically display historical data via canned or custom-built views. It can also perform calculations on data or feed other Amazon data analysis services for more sophisticated study.
  • Compliance and security. CloudWatch integrates with AWS identity and access management to enforce role-based access controls on CloudWatch data and automatically encrypt stored data.

CloudWatch vs. Datadog and New Relic

CloudWatch certainly has better functionality now than it did 10 years ago. However, enterprises with sizable legacy infrastructure, as well as hybrid and multi-cloud applications, will find it lacking, with some notable deficiencies.

For example, CloudWatch can't monitor non-AWS cloud services or user-managed infrastructure. In addition, the tool has limited to no visibility into telemetry from the OS stack and running applications.

For more advanced analysis and automation, CloudWatch also requires users to export data to other services and to develop custom scripts, Lambda functions, business intelligence routines or machine learning models.

When you compare CloudWatch vs. Datadog and New Relic, both third-party tools offer capabilities that CloudWatch lacks. Datadog and New Relic provide unified infrastructure and application performance monitoring suites that include an extensive set of service integrations, pre-built dashboards and data analysis models. Along with competitors such as AppDynamics, both Datadog and New Relic are good options for organizations that want to add AWS infrastructure into a multi-platform monitoring and alerting system.

Datadog vs. New Relic pricing

Just as their AWS monitoring features differ, Datadog and New Relic have different pricing models, as well. For example, Datadog, at the time of publication, costs $0.03 per host per hour for AWS infrastructure monitoring, while New Relic costs $149 per month for 125,000 compute units (CUs). According to New Relic, CU-based pricing calculates the monthly charge based on the size of the host -- compute power and memory -- running New Relic, and the number of hours it connects to New Relic during the month.

For more details on Datadog and New Relic pricing, visit the AWS Marketplace.

As the table below illustrates, Datadog and New Relic share many features and have the same general capabilities, but differ in the emphasis and usability of their implementations. In general, the strength of each stems from its origins: Datadog started as an infrastructure monitoring system that added APM features over time, while New Relic began as an APM platform that added infrastructure monitoring to round out its feature set.

Thus, development and DevOps teams might prefer New Relic, while IT ops teams might be more comfortable with Datadog. The following table summarizes the tools' core features:

Datadog vs. New Relic comparison chart

*According to Datadog and New Relic websites, at time of publication

Take advantage of free trial periods before you commit to one of these tools. Organizations that don't need to integrate legacy on-premises systems or workloads on other clouds into a single monitoring interface should start with Amazon CloudWatch, and only investigate third-party alternatives when they hit limits or can't access a required capability.

Next Steps

AWS monitoring best practices extend beyond CloudWatch

Compare Grafana vs. Datadog for IT monitoring

Learn how New Relic works, and when to use it for IT monitoring

Dig Deeper on AWS infrastructure

App Architecture
Cloud Computing
Software Quality