OQS algorithm performance visualizations

Disclaimer

These pages visualize measurements taken by the now-defunct OQS profiling project. This project is not currently maintained, and these measurements are not up to date. The data contained here are not representative of the current liboqs main branch, nor are they representative of the liboqs main branch at the time they were collected. The liboqs version and commit for which they were collected can be seen by viewing the raw data, which is linked on each page below.

Benchmarking

1. KEM algorithms runtime (liboqs speed_kem)
2. SIG algorithms runtime (liboqs speed_sig)
3. KEM algorithms memory use (liboqs test_kem)
4. SIG algorithms memory use (liboqs test_mem_sig)
5. KEM/SIG OpenSSL performance (openssl speed)
6. TLS handshakes performance (openssl s_time)

Additional information

Algorithms labelled -ref are reference (C-code only) implementations compiled without special CPU feature support. Algorithms without this moniker are compiled with all CPU extensions available but with code portability/distributability features switched on (see here for details about this option). Algorithm names ending in -noport represent measurements taken without portability/distributability enabled (i.e., such code may crash when executed on machines without suitable performance-enhancing CPU features).

These graphs will be extended over time to other CPU types and cloud infrastructures.

We are grateful to Amazon Web Services for supplying compute-time credits for our benchmarking platform.