Today, AWS announces the general availability of runtime metrics support in Amazon CloudWatch Application Signals, an OpenTelemetry (OTel) compatible application performance monitoring (APM) feature in CloudWatch. You can view runtime metrics like garbage collection, memory usage, and CPU usage for your Java or Python applications to troubleshoot issues such as high CPU utilization or memory leaks, which can disrupt the end-user experience.
Application Signals simplifies troubleshooting application performance against key business or service level objectives (SLOs) for AWS applications. Without any source code changes, Application Signals collects traces, application metrics(error/latency/throughput), logs and now runtime metrics to bring them together in a single pane of glass view.
Runtime metrics enable real-time monitoring of your application’s resource consumption, such as memory and CPU usage. With Application Signals, you can understand whether anomalies in runtime metrics have any impact on your end-users by correlating them with application metrics such as error/latency/throughput. For example, you will be able to identify if a service latency spike is a result of an increase in garbage collection pauses by viewing these metric graphs side by side. Additionally you will be able to identify thread contention, track memory allocation patterns, and pinpoint memory or CPU spikes that may lead to application slowdowns or crashes, impacting end user experience.
Runtime metrics support is available in all regions Application Signals is available in. Runtime metrics are charged based on Application Signals pricing, see Amazon CloudWatch pricing.
To learn more, see documentation to enable Amazon CloudWatch Application Signals.
Source:: Amazon AWS