OpenSSL Performance Benchmark

In his presentation, Martin Schmatz raised important concerns about Dilithium's (ML-DSA) k-values and the rejection sampling retry mechanism. This is a critical consideration for stress testing post-quantum signature algorithms.

We've implemented dedicated testing for this: See the Post-Quantum Cryptography page for detailed ML-DSA rejection sampling analysis, including:

• Coefficient of Variation (CV%) - Measures signing time variance
• P99, P99.9, P99.99 latencies - Tail latency analysis for capacity planning
• Outlier detection - Operations taking >2× the mean time (indicating many retries)

The benchmark runs for 90 seconds to collect ~108,000 samples for statistically robust P99.99 measurements.

What This Chart Shows: This benchmark measures AES-256-GCM encryption throughput across different block sizes (16 bytes to 8KB) to reveal how cryptographic operations scale with data size.

Key Insights:

• Small blocks (16-64 bytes) stress initialization overhead - each encryption requires Provider setup, key scheduling, and context creation
• Medium blocks (256 bytes - 1KB) show the transition point where throughput begins to increase
• Large blocks (8KB+) achieve maximum throughput by amortizing initialization costs across more data
• The gap between versions reveals Provider architecture overhead in OpenSSL 3.x compared to 1.1.1w

Real-World Impact: Applications encrypting small messages (e.g., individual database fields, IoT sensor data) will see much lower throughput than bulk encryption (file encryption, large API payloads).