Load tests verify that an API or service performs acceptably under expected and peak traffic conditions. Unlike functional tests that validate correctness, load tests validate capacity: response time under concurrent requests, throughput at various traffic levels, error rates as load increases, and behavior at saturation. Load tests answer the critical question: will the system handle production traffic without degrading performance or failing? They are most valuable in pre-production environments during capacity planning exercises and after significant architectural changes that might affect performance characteristics. Running load tests before deployment prevents performance issues from reaching production.

The generator takes an OpenAPI specification and produces a load test targeting the specified endpoints. Required inputs include the target endpoint, expected concurrent user count, request rate (requests per second or per minute), test duration, and authentication configuration. Optional inputs include custom headers, request body examples, and ramp-up period configuration for gradual load increases. The generator produces a deterministic load test that runs against the configured traffic profile. The execution model spawns concurrent virtual users (goroutines), each repeatedly executing the scenario for the configured duration while distributing the target request rate across all virtual users. Same inputs produce the same load test every time, making load test configurations version-controllable and reproducible across environments.

Load tests belong in CI/CD pipelines as performance gates, not just as one-time exercises. The standard pattern: after deployment to a staging environment, the load test runs against the deployed service and validates that response times and error rates remain within configured thresholds. If performance degrades beyond the threshold, the pipeline fails and the deployment does not promote to production. This catches performance regressions before they reach production, not after users report them. Automated load testing in pipelines transforms performance validation from an occasional exercise into a continuous gate that prevents performance degradation from ever reaching production.

Developers often search for the distinctions between load, stress, and spike testing. Load testing validates performance under expected traffic levels, testing that the system handles normal and peak production load without degradation. Stress testing pushes the system beyond expected capacity to find the breaking point, identifying the maximum load the system can handle before failures occur. Spike testing validates behavior under sudden, sharp traffic increases, testing that the system handles traffic surges without failing. Skyramp's default load test generator targets the load testing use case: sustained traffic at expected or slightly above-expected levels over a defined period. For teams needing stress or spike test profiles, the same base configuration can be adapted by modifying the traffic parameters.