Operations | Monitoring | ITSM | DevOps | Cloud

Sometimes Docker behaves in ways that catch you off guard—containers don’t start as expected, images pause during pull, or networking takes longer than usual to respond. In those moments, the Docker daemon logs are your best reference point. These logs capture exactly what the Docker engine is doing at any given time. They give you a running account of system state, performance signals, and events that help you understand what’s happening beneath the surface.

Are your Java applications running at their optimal performance, or is there room for improvement to make them faster and more efficient? With so many services depending on Java, keeping applications responsive and reliable is a core part of modern software engineering. This blog walks you through the leading Java Application Performance Monitoring (APM) tools, with a clear comparison to help you choose the right option for your needs.

It’s 3 AM. API latency just spiked from 200ms to 2s. Alerts are firing, and users are frustrated. You SSH into the first server: top, free -h, iostat — nothing unusual. On to the next host. And the next. That’s how most of us learned to debug. The tools worked, and we got good at using them. But as infrastructure became distributed and dynamic, this approach started to break down. Modern monitoring needs more than SSH and top. It needs unified telemetry.

Application Performance Monitoring (APM) helps you understand how your software runs in production. When you track the right metrics, you see how requests move through your system, where slowdowns happen, and how resources are being used. With this knowledge, you can spot issues early and keep your applications reliable for your users. In this blog, we discuss the key APM metrics to monitor, grouped into categories, and why each one matters for performance and user experience.

Microservices make it tough to understand how applications behave end-to-end. Most teams already rely on an Application Performance Monitoring (APM) tool to track system health. But as requests move across many services, you also need distributed tracing. Jaeger gives you that visibility. The real value comes from connecting the two. Instead of running APM and Jaeger in silos, you can combine their strengths, metrics from your APM, and traces from Jaeger, to get a clearer view of performance.

You've got Prometheus running in one cluster — maybe a dev environment, a single EKS cluster, or a proof-of-concept setup. The configuration is straightforward: node_exporter on a few EC2 instances, some service discovery for pods, and a single Prometheus server scraping everything. Storage is local, retention is 15 days, and you can keep all the default recording rules without worrying about costs.

Modern applications don’t process everything inside the request/response path. To keep APIs responsive, time-consuming work like image resizing, payment processing, or data syncs is moved into background queues. Workers then pick up these asynchronous jobs and run them outside the main thread. Asynchronous job monitoring is the practice of tracking these background tasks: Without this visibility, background workers become a blind spot.

In Kubernetes, applications are constantly changing — new pods start, old ones shut down, workloads shift across nodes. The challenge is making sure that different parts of your system, and even external clients, can still find each other when the actual locations keep moving. That’s what service discovery handles. It provides a stable way for applications to connect and communicate, no matter where they’re running or how often the underlying infrastructure changes.

Background jobs handle the critical work that happens outside the request path: processing payments, sending emails, generating reports, syncing data. They keep applications running smoothly, but the signals they produce look different from API endpoints. Most teams start with queue metrics—how many jobs are waiting and how quickly they complete. These metrics provide the foundation, but job health extends beyond throughput.

A service catalog gives teams a shared view of their systems—what services exist, who owns them, how dependencies are structured, and the SLAs that guide expectations. It’s an important part of development infrastructure because it helps everyone speak the same language about services. Service catalog observability builds on that foundation.

When a payment service runs across 12 pods — each serving different customer segments — and an authentication layer spans three namespaces, performance issues can originate in both the application code and the orchestration layer. The challenge is linking request-level performance data with what’s happening inside the cluster: container CPU limits, pod scheduling decisions, and node-level events.

A Kubernetes cluster can generate more than 1,400 metrics out of the box. That’s a lot of numbers to sift through, especially when you’re troubleshooting a production slowdown in the middle of the night. The key is knowing which metrics tell you the most, with the least noise. These are the signals worth paying attention to when you need answers fast.

APM tracing records the complete execution path of a request as it travels through your system, including database queries, external API calls, cache lookups, message queue events, and inter-service requests. Each step is captured with precise start and end timestamps, duration, and context such as service name, operation name, and relevant attributes. This lets you pinpoint where latency or errors originate without piecing together metrics and logs manually.

The OpenTelemetry Gateway (OTel Gateway) is a centralized service that collects, processes, and routes telemetry data—metrics, traces, and logs—across your infrastructure. In a typical setup, each service pushes telemetry directly to an observability backend. While this approach works well for small environments, it becomes increasingly difficult to manage as systems grow.