Performance

Rift is designed for high performance, delivering 2-250x better throughput than Mountebank.

Benchmark Summary

Test configuration: 15s duration, 50 concurrent connections, 2 CPUs, 1GB RAM per service.

Standout Results

Feature Mountebank Rift Speedup
JSONPath Predicates 107 RPS 26,500 RPS 247x faster
XPath Predicates 169 RPS 28,700 RPS 170x faster
API Stub (Last Match) 290 RPS 22,000 RPS 76x faster
404 Handling 297 RPS 22,600 RPS 76x faster
Complex Predicates 904 RPS 29,300 RPS 32x faster
JSON Body Matching 1,022 RPS 24,000 RPS 24x faster
Template Responses 1,349 RPS 28,100 RPS 21x faster
Simple Health Check 1,914 RPS 39,100 RPS 20x faster
High Concurrency 1,830 RPS 29,700 RPS 16x faster

Performance by Category

Core Functionality

Test Mountebank (RPS) Rift (RPS) Speedup
Health Check 1,914 39,127 20x
Ping/Pong 1,763 36,391 21x
List Imposters 6,361 29,021 4.5x
Get Imposter 396 884 2.2x

API Stub Matching (500 stubs)

Test Mountebank (RPS) Rift (RPS) Speedup
First Stub Match 1,865 33,422 18x
Middle Stub Match 589 25,512 43x
Last Stub Match 290 22,042 76x
No Match (404) 297 22,664 76x

JSONPath Predicates

Test Mountebank (RPS) Rift (RPS) Speedup
First Match 107 26,583 247x
Middle Match 129 28,751 221x
Last Match 124 27,070 218x

XPath Predicates

Test Mountebank (RPS) Rift (RPS) Speedup
First Match 169 28,745 170x
Middle Match 168 27,234 161x
Last Match 173 27,248 157x

Why Is Rift Faster?

Architecture Comparison

Aspect Mountebank Rift
Language Node.js (JavaScript) Rust
Concurrency Single-threaded event loop Multi-threaded (Tokio)
Memory Model Garbage collected Zero-copy, no GC
Regex Engine JavaScript RegExp Rust regex crate
JSON Parsing JavaScript JSON serde_json (SIMD)
Stub Matching Linear scan Optimized matching

Key Optimizations

  1. Native Code: Rust compiles to native machine code, avoiding interpreter overhead.

  2. Async I/O: Tokio runtime provides efficient async networking with work-stealing scheduler.

  3. Zero-Copy Parsing: serde_json parses JSON without unnecessary allocations.

  4. Efficient Regex: Rust’s regex crate uses finite automata for O(n) matching.

  5. Connection Pooling: Reuses connections to upstream services.

  6. Thread Pool: Dedicated workers for script execution.

Performance Characteristics

Latency Distribution

Scenario Mountebank P99 Rift P99
Simple stub 50ms 2ms
Complex predicate 150ms 5ms
JSONPath match 800ms 3ms
Under load (200 conn) 300ms 15ms

Throughput Scaling

Rift maintains consistent throughput regardless of:

  • Stub count (500+ stubs with minimal degradation)
  • Stub position (first vs last stub match)
  • Predicate complexity

Mountebank shows linear degradation as stub count increases.

Running Benchmarks

Prerequisites 

# Install hey (HTTP load generator)
brew install hey  # macOS
# or
go install github.com/rakyll/hey@latest

Run Benchmark Suite 

cd tests/benchmark

# Start services
docker compose up -d --build

# Run benchmarks
./scripts/run-benchmark.sh

# View results
cat results/BENCHMARK_REPORT.md

# Cleanup
docker compose down -v

Custom Configuration 

# Longer duration, more connections
DURATION=60s CONNECTIONS=100 ./scripts/run-benchmark.sh

# Quick smoke test
DURATION=10s CONNECTIONS=20 ./scripts/run-benchmark.sh

Optimization Tips

For Maximum Throughput

  1. Use specific predicates - equals is faster than matches
  2. Order stubs by frequency - Most-matched stubs first
  3. Avoid unnecessary behaviors - Each behavior adds overhead
  4. Use native formats - JSON body predicates are faster than string matching

For Lowest Latency

  1. Minimize stub count - Fewer stubs = faster matching
  2. Use simple responses - Static is responses are fastest
  3. Avoid injection - JavaScript execution adds latency
  4. Enable connection pooling - Reuse upstream connections

Resource Allocation 

# Recommended for high throughput
resources:
  requests:
    cpu: 1000m
    memory: 256Mi
  limits:
    cpu: 2000m
    memory: 512Mi

Comparison with Alternatives

Tool Language Typical RPS Best For
Rift Rust 20,000-40,000 High-performance mocking
Mountebank Node.js 500-2,000 Feature-rich service virtualization
WireMock Java 1,000-5,000 Java ecosystem integration
MockServer Java 1,000-3,000 Contract testing

Rift provides 10-100x better performance while maintaining Mountebank compatibility.