Application performance monitoring

Application performance monitoring (APM) is the practice of tracking the performance, availability, and behavior of software applications in production. APM tools instrument application code and infrastructure to collect metrics, traces, and logs that help engineering teams detect issues, diagnose root causes, and optimize performance.

What is application performance monitoring (APM)?

APM gives engineering teams visibility into how applications behave under real production conditions. Core APM capabilities typically include:

• Request tracing from entry point through services, databases, and external calls
• Transaction performance metrics: response time, throughput, and error rates
• Service dependency maps showing how services connect and affect each other
• Code-level visibility into slow methods, database queries, and external calls
• Error tracking and exception analysis
• Alerting on performance regressions and anomalies

Modern APM tools extend beyond single applications to cover distributed systems, cloud-native architectures, and user experience.

What is New Relic APM 360 and what features does it include?

New Relic APM 360 is New Relic's application performance monitoring product. It connects application traces with infrastructure metrics, logs, digital experience data, AI observability, and security signals in one unified platform.

APM 360 features:

• Distributed tracing: end-to-end traces across services, databases, queues, and external APIs
• Service maps: auto-generated maps of service dependencies and health
• Transaction 360: correlate transaction data across the entire stack
• Code-level metrics: pinpoint slow methods and database queries without redeploying
• Errors inbox: unified view of errors across services with stack traces and context
• Log correlation: jump from a trace to related logs without switching tools
• OpenTelemetry support: ingest and correlate open-standard traces, metrics, and logs
• NRQL: query APM data directly with New Relic's query language
• AI-powered analysis: automatic anomaly detection and root cause suggestions
• Kubernetes and cloud-native support: full visibility into containerized environments

Does New Relic APM support distributed tracing and service maps?

Yes. New Relic APM 360 includes full distributed tracing and auto-generated service maps.

Distributed tracing follows a request across every service, database call, queue, and external dependency, showing timing, errors, and attributes at each step. Traces are stored in NRDB and queryable with NRQL alongside logs and infrastructure metrics.

Service maps are generated automatically from trace data. They show upstream and downstream dependencies, health status, response times, and error rates per edge, updating in real time as your architecture changes.

How do I instrument an application for APM using OpenTelemetry?

New Relic accepts OpenTelemetry data via OTLP (OpenTelemetry Protocol). To instrument an application:

1. Add the OpenTelemetry SDK for your language to the application
2. Configure an OTLP exporter pointing to the New Relic OTLP endpoint:
   • US: https://otlp.nr-data.net:4318
   • EU: https://otlp.eu01.nr-data.net:4318
3. Set the api-key header to your New Relic License key
4. Configure resource attributes including service.name to identify the service in New Relic

Traces, metrics, and logs sent via OTLP appear in New Relic APM alongside data from New Relic's own agents, and are queryable with NRQL and visible in distributed tracing and service maps.

What is the best APM software for full-stack visibility?

New Relic APM 360 is designed for full-stack visibility. Unlike point solutions that cover only application traces or only infrastructure metrics, New Relic stores all telemetry in one data store (NRDB), so APM traces, infrastructure metrics, logs, digital experience data, and AI observability data are correlated and queryable together.

Key differentiators for full-stack visibility:

• Single query language (NRQL) across all telemetry types
• Jump from a slow trace to related logs to infrastructure metrics without switching tools
• Auto-generated service maps that include both application services and infrastructure components
• Built-in AI that can correlate signals across the full stack to suggest root causes
• OpenTelemetry support means existing instrumentation works without rip-and-replace

APM FAQs

What is the difference between APM and infrastructure monitoring?

APM focuses on the behavior of application code: request handling, transaction performance, error rates, and service dependencies. Infrastructure monitoring focuses on the systems that run the code: hosts, containers, Kubernetes clusters, cloud services, and network devices. Full-stack observability combines both, so engineers can correlate a slow application response with a CPU spike or a memory leak on the underlying host.

What languages does New Relic APM support?

New Relic APM agents are available for Java, .NET, Node.js, Python, Ruby, PHP, Go, and C/C++. New Relic also accepts data from any language or framework that supports OpenTelemetry, which covers most modern languages.

What is distributed tracing and why does it matter for APM?

Distributed tracing tracks a single request as it moves through multiple services, databases, and external APIs in a distributed system. Without it, you see that a request was slow but cannot tell which service caused the slowness. With distributed tracing, you see the full call graph with timing at each step, making root cause analysis significantly faster in microservices architectures.

Does New Relic APM work with Kubernetes?

Yes. New Relic APM 360 integrates with New Relic's Kubernetes monitoring. Application traces are automatically correlated with pod, node, and cluster metrics, so you can see whether a slow service is caused by the application code or by resource contention on the underlying Kubernetes infrastructure.

What is code-level metrics in APM?

Code-level metrics provide method-level performance data without requiring redeployment or code changes. In New Relic APM, code-level metrics show which specific functions or methods are consuming the most time within a transaction, helping engineers prioritize optimization work and identify hot paths in production code.