Benchmark Results
This document presents real-world benchmarking results for Solvix, demonstrating its performance, stability, and efficiency compared to other HTTP clients.
Why Benchmarking Matters
Benchmarking helps validate:
- Performance under load
- Stability over time
- Efficiency vs competitors
- Real-world production readiness
Test Environment
- Runtime: Node.js v24
- Duration per test: ~2 seconds
- Hardware: Standard development machine
- Tooling: Tinybench
Basic Throughput Test
Result (Solvix)
- Mean latency: ~0.001 ms
- Throughput: ~1,000,000 req/sec
This demonstrates extremely low overhead.
Comparison
| Library | Avg Latency | Throughput |
|---|---|---|
| Solvix | ~0.001 ms | ~1M req/s |
| Native fetch | ~0.007 ms | ~130k req/s |
| Axios | ~0.021 ms | ~47k req/s |
Real Network Test
Solvix
- Avg latency: ~100–117 ms
- Stable under repeated runs
Observations
- Minimal overhead vs network time
- Consistent performance
- Efficient retry + pipeline handling
Memory Stability Test
Results
- Initial heap: ~7 MB
- After 10k requests: ~10 MB
- Stable across multiple rounds
Conclusion
No memory leaks detected.
Concurrency Test
Solvix handles high concurrency efficiently using:
- Internal queue
- Rate limiter
- Deduplication
Result
- Stable throughput
- No request flooding
- Controlled execution
Retry Storm Test
Simulated failing requests with retries.
Result
- Controlled retry behavior
- No exponential explosion
- System stability maintained
Token Refresh Stampede Test
Multiple requests triggering token refresh simultaneously.
Result
- Only 1 refresh executed
- All requests reused the same token
This proves stampede protection works correctly.
Deduplication Test
Multiple identical requests executed concurrently.
Result
- Only 1 network call executed
- Others reused the same promise
Bundle Size
- Minified: ~23 KB
- Gzip: ~7.8 KB
Conclusion
- Lightweight
- Production-friendly
Key Takeaways
- Extremely low overhead
- High throughput
- Memory stable
- Concurrency safe
- Production-ready resilience
Summary
Solvix is optimized for:
- Performance
- Reliability
- Scalability
It outperforms traditional HTTP clients in controlled benchmarks and maintains stability in real-world scenarios.