Performance Testing Tools is a dense, practical directory for engineers
choosing tools for load, stress, protocol, and performance testing.
How to read the directory
The original PerformanceTestingTools diagram used a legend to call out
the author's personal picks and generally recommended tools. This
directory keeps that signal visible: ♥ personal pick marks a personal recommendation, while ✓ general pick marks a broadly useful choice.
Mini load profiles are deterministic visual fingerprints, not measured
benchmarks. They are generated from each tool's catalog attributes,
not measured performance.
Profile archetypes
RampGradual rise for protocol/API-oriented tools.
StepStaged plateaus common for enterprise suites.
SpikeConcentrated peak common for micro-benchmark CLIs.
SoakSustained oscillating load common for open-source engines.
BurstRepeated bursts common for cloud/commercial services.
Maintained with evidence
Tool records include official links, ownership, licensing, deployment,
protocols, and lifecycle status where available. Renamed and
discontinued products remain visible so historical references are not
silently lost.
Every record starts from the vendor’s own documentation, pricing page and source repository, then keeps the links so any claim can be re-checked.
Volatile fields — pricing, latest release, feature lists — are refreshed from an evidence-gathering pass and stamped with the date they were verified; the date is printed on every tool page.
Comparisons and alternatives are generated from the recorded fields, so they say what the catalog can support and nothing more.
Nothing here is a benchmark. There are no ratings, no scores, and no measured throughput numbers; the load-profile sketches are deterministic drawings from a tool’s attributes, not results.
Discontinued tools stay listed, with a successor when one is verified, so older references still resolve.
Corrections are public: anything wrong can be fixed through a GitHub issue.