Day 34 of 60
·
Load, chaos, durability
Load testing
A passing single-request test guarantees nothing about behavior at peak. Load testing is the cheapest way to learn about your latency cliff before your customers do.
ProblemPerformance fails under realistic traffic; and you find out at launch.
How it works
Replay synthetic traffic at target volumes; record latency, throughput, error rate. Run nightly or pre-release on representative environments.
What it catches
Latency cliffs, connection-pool exhaustion, GC pauses, cache misses, capacity-planning errors.
Tools
k6 · OSS Locust · OSS Gatling · OSS JMeter · OSS
Verdict by project size
Small
Skip
Medium
Rec
Large
Must
Extra-large
Must
Cost
| Project size | Setup | Maint / mo | Tool / mo | CI / run |
|---|---|---|---|---|
| Small <10k LOC | 1d | 1h | $0 | , |
| Medium 10–100k LOC | 3d | 4h | $0 | , |
| Large 100k–1M LOC | 10d | 20h | $500 | , |
| Extra-large >1M LOC | 30d | 80h | $5k | , |
Setup = engineer-days to first useful run ·
Maint = engineer-hours / month at steady state ·
Tool = out-of-pocket $ / month ·
CI = minutes added (or saved) per pipeline run
Lifecycle & ownership
When in lifecycle
Test Release Operate
Per release · Runs before promotion to production.
Who owns it
SRE / DevOps / Platform
CI/CD, observability, reliability
Collaborates with: Developer
Reference implementations
-
k6 examples
Executable load-testing scenarios for APIs, browsers, and CI.
-
Gatling demos
Load-test scenarios and recorder fixtures from the Gatling project.
-
Locust examples
Python load-testing examples for HTTP and user behavior modeling.
Quick check
A passing single-request test guarantees…
One question. Pick the best answer. Your streak is saved locally on this device.
Save the lesson
Download SVG ↓
Screenshot for a 1:1, drop it in Slack, or download the SVG.