Skip to content

Server Monitoring

A load test that shows slow response times tells you something is wrong. Server monitoring tells you what. CPU pegged at 100%, memory exhausted, disk I/O saturated, network bandwidth maxed: each leaves a different fingerprint, and each calls for a different fix. Without server-side metrics you're guessing.

Load Tester collects host metrics two ways. Through its own monitoring agent installed on the server, or through CloudWatch when the server runs on AWS. Both feed the same Servers View and the same correlation logic the Analytics Dashboard uses to line up response-time spikes against the resource that ran out first.

Server monitoring is included at every license tier. There's nothing to buy separately.

In This Section

  • Introduction: What server monitoring measures, why it matters for load testing, and how it fits into the broader analysis workflow.
  • Basic Monitoring: The simplest case. Pointing Load Tester at a server, picking metrics, and seeing them live in the Servers View.
  • CloudWatch Monitoring: AWS-hosted targets where CloudWatch is the source of truth instead of an installed agent.
  • Monitoring Agent: Installing the cross-platform agent on Windows, macOS, or Linux servers, with the configuration knobs that actually matter.
  • Metrics & Counters: The specific OS counters Load Tester collects, what each one means, and which ones are usually load-bearing for diagnosing performance issues.
  • Through Firewalls: Configuring the agent and Load Tester to communicate across network boundaries without opening more ports than necessary.
  • Performance Checklist: A short, opinionated list of what to monitor on a typical web server. Use it as a starting point and adjust for your stack.

For first-time setup, Introduction then Basic Monitoring. For AWS targets, skip directly to CloudWatch Monitoring. For everything else, Monitoring Agent is the install path.