PQ-WB-KEM: White-Box ML-KEM-768 1.0
PQ-WB-KEM: Toward a White-Box Construction of ML-KEM-768 with Arithmetic Masking for M2M Communications - Mathematics (MDPI)
Loading...
Searching...
No Matches
wallclock_ci.py File Reference

Aggregate the platform-dependent (wall-clock) benchmarks over N runs -> mean +/- 95% CI. More...

Go to the source code of this file.

Namespaces

namespace  wallclock_ci
 

Functions

 wallclock_ci.detect_env ()
 Return a short environment tag: 'windows' or 'wsl' (or 'linux').
 
 wallclock_ci.main (n_runs=20)
 
 wallclock_ci.oqs_available ()
 
 wallclock_ci.stats (xs)
 
 wallclock_ci.tval (n)
 

Variables

 wallclock_ci.HERE = os.path.dirname(os.path.abspath(__file__))
 
 wallclock_ci.RESULTS = os.path.join(os.path.dirname(os.path.dirname(HERE)), "results")
 
dict wallclock_ci.TVAL
 

Detailed Description

Aggregate the platform-dependent (wall-clock) benchmarks over N runs -> mean +/- 95% CI.

Wall-clock timings vary run-to-run and across liboqs builds/CPUs; a single run is not a robust figure. This driver repeats each wall-clock measurement N times and reports mean +/- 95% CI (with per-run samples) so the paper can cite an interval rather than a single-run point. Deterministic experiments (masking chi^2, algebraic 768-recovery, DCA correlation, footprint, NTT) reproduce exactly and are excluded.

Two measurement paths, chosen for cost:

  • liboqs baseline (keygen/encaps/decaps): subprocess the full benchmark N times.
  • masking overhead + algebraic countermeasure overhead: call the timed routines in-process N times (the slow deterministic recovery is NOT repeated).
Author
Ugur Coruh

Definition in file wallclock_ci.py.