ngcc.dev

Performance measurements › arm_1 › kem-16 · system: x86_1 · arm_1

kem-16 HARE — performance on AArch64 (system arm_1)

Independent measurement following the structure of the NICCS ARM self-assessment guide, §3.5 (1)–(7). Not a submitter self-assessment.

1. Basic information

2. Assessment environment

item value
processor Qualcomm Oryon (CPU 2, one core)
machine ASUS Vivobook S 15
clock fixed 2.71 GHz (governor performance, minimum = maximum), boost off, SMT none
memory 30562 MiB
OS / kernel Ubuntu 26.04.1 LTS / 7.0.0-34-generic
compiler / build tool gcc (Ubuntu 15.2.0-16ubuntu1) 15.2.0 / cmake version 4.2.3
campaign start / end (UTC) 2026-09-28T15:05:46 / 2026-09-29T08:43:52

3. Functional testing (KAT)

Each library was checked against the submitted KAT vectors (SHA-256 manifest kem-16/kat.sha256 in the harness) before timing.

instance build flags KAT
HARE-128-kr guide PASS
HARE-256-kr guide PASS
HARE-384-kr guide PASS
HARE-512-kr guide PASS

4. Performance

instance operation mean cycles mean time ops/s median trial n (trials × iters)
HARE-128-kr keygen 4.13 M 1.53 ms 652 1.53 ms 1905 (5 × 381)
HARE-128-kr enc 9.54 M 3.54 ms 283 3.54 ms 875 (5 × 175)
HARE-128-kr dec 14.16 M 5.25 ms 190 5.25 ms 600 (5 × 120)
HARE-256-kr keygen 15.89 M 5.9 ms 170 5.89 ms 530 (5 × 106)
HARE-256-kr enc 34.95 M 13 ms 77.1 13 ms 245 (5 × 49)
HARE-256-kr dec 53.12 M 19.7 ms 50.7 19.7 ms 165 (5 × 33)
HARE-384-kr keygen 47.23 M 17.5 ms 57.1 17.5 ms 180 (5 × 36)
HARE-384-kr enc 100.50 M 37.3 ms 26.8 37.3 ms 100 (5 × 20)
HARE-384-kr dec 150.39 M 55.8 ms 17.9 55.8 ms 100 (5 × 20)
HARE-512-kr keygen 100.69 M 37.4 ms 26.8 37.4 ms 100 (5 × 20)
HARE-512-kr enc 210.06 M 77.9 ms 12.8 77.9 ms 100 (5 × 20)
HARE-512-kr dec 316.46 M 117 ms 8.52 117 ms 100 (5 × 20)

5. Resource consumption

Static memory is approximated by the library's loadable ELF segments; peak memory is the process high-water mark (VmHWM) including the benchmark driver and libc, so both are upper-bound proxies rather than isolated algorithm memory.

instance operation static: ELF image (bytes) baseline RSS peak RSS
HARE-128-kr keygen 46944 1448 KiB 1576 KiB
HARE-128-kr enc 46944 1516 KiB 1584 KiB
HARE-128-kr dec 46944 1532 KiB 1600 KiB
HARE-256-kr keygen 51664 1464 KiB 1684 KiB
HARE-256-kr enc 51664 1644 KiB 3756 KiB
HARE-256-kr dec 51664 1684 KiB 1752 KiB
HARE-384-kr keygen 52304 1492 KiB 1836 KiB
HARE-384-kr enc 52304 1844 KiB 1916 KiB
HARE-384-kr dec 52304 1932 KiB 2000 KiB
HARE-512-kr keygen 52824 3488 KiB 4032 KiB
HARE-512-kr enc 52824 4080 KiB 4144 KiB
HARE-512-kr dec 52824 4232 KiB 4296 KiB

6. Transmission and storage overhead

External public-key, ciphertext and signature sizes follow the curated performance/external_sizes.csv catalog; secret-key and shared-secret lengths remain API figures. See the size audit for disagreements.

instance public key secret key ciphertext shared secret
HARE-128-kr 2629 2677 4688 16
HARE-256-kr 6580 6676 11790 32
HARE-384-kr 13157 13301 23693 48
HARE-512-kr 21812 22004 39294 64

Symmetric primitives

Survey verdict: ICCS-only — pseudoXOF only (recomputes prefix per call); lives under _shared/

Share of each operation spent in the ICCS placeholder hash functions and in the ICCS DRNG (reference build, measured in the same run with link-time wrappers):

instance operation pseudohash/XOF/sm3 DRNG calls per operation
HARE-128-kr keygen 4.5% 0.1% drng 1, pseudoXOF 6.39
HARE-128-kr enc 3.2% 0.1% drng 2, pseudoXOF 6
HARE-128-kr dec 3.0% 0.0% pseudoXOF 8
HARE-256-kr keygen 2.4% 0.0% drng 1, pseudoXOF 6.55
HARE-256-kr enc 1.7% 0.0% drng 2, pseudoXOF 6
HARE-256-kr dec 1.6% 0.0% pseudoXOF 8
HARE-384-kr keygen 1.7% 0.0% drng 1, pseudoXOF 7.58
HARE-384-kr enc 1.4% 0.0% drng 2, pseudoXOF 6
HARE-384-kr dec 1.6% 0.0% pseudoXOF 9
HARE-512-kr keygen 2.4% 0.0% drng 1, pseudoXOF 7.53
HARE-512-kr enc 1.6% 0.0% drng 2, pseudoXOF 6
HARE-512-kr dec 1.6% 0.0% pseudoXOF 9

7. Raw evidence index

Paths are relative to performance/data/arm_1/ in the harness; each JSON file records its commands, environment and trials.

instance item file
HARE-128-kr KAT log (sha256 6e3ec02996e97149…) kat/kem-16/HARE-128-kr.log
HARE-128-kr timing dec records/kem-16/HARE-128-kr__dec.json
HARE-128-kr timing enc records/kem-16/HARE-128-kr__enc.json
HARE-128-kr timing keygen records/kem-16/HARE-128-kr__keygen.json
HARE-128-kr hash profile dec profile/kem-16/HARE-128-kr__dec.json
HARE-128-kr hash profile enc profile/kem-16/HARE-128-kr__enc.json
HARE-128-kr hash profile keygen profile/kem-16/HARE-128-kr__keygen.json
HARE-256-kr KAT log (sha256 9c49f57f0f59e50d…) kat/kem-16/HARE-256-kr.log
HARE-256-kr timing dec records/kem-16/HARE-256-kr__dec.json
HARE-256-kr timing enc records/kem-16/HARE-256-kr__enc.json
HARE-256-kr timing keygen records/kem-16/HARE-256-kr__keygen.json
HARE-256-kr hash profile dec profile/kem-16/HARE-256-kr__dec.json
HARE-256-kr hash profile enc profile/kem-16/HARE-256-kr__enc.json
HARE-256-kr hash profile keygen profile/kem-16/HARE-256-kr__keygen.json
HARE-384-kr KAT log (sha256 c8b7e261ad1aea7b…) kat/kem-16/HARE-384-kr.log
HARE-384-kr timing dec records/kem-16/HARE-384-kr__dec.json
HARE-384-kr timing enc records/kem-16/HARE-384-kr__enc.json
HARE-384-kr timing keygen records/kem-16/HARE-384-kr__keygen.json
HARE-384-kr hash profile dec profile/kem-16/HARE-384-kr__dec.json
HARE-384-kr hash profile enc profile/kem-16/HARE-384-kr__enc.json
HARE-384-kr hash profile keygen profile/kem-16/HARE-384-kr__keygen.json
HARE-512-kr KAT log (sha256 5378649b0f3e29f7…) kat/kem-16/HARE-512-kr.log
HARE-512-kr timing dec records/kem-16/HARE-512-kr__dec.json
HARE-512-kr timing enc records/kem-16/HARE-512-kr__enc.json
HARE-512-kr timing keygen records/kem-16/HARE-512-kr__keygen.json
HARE-512-kr hash profile dec profile/kem-16/HARE-512-kr__dec.json
HARE-512-kr hash profile enc profile/kem-16/HARE-512-kr__enc.json
HARE-512-kr hash profile keygen profile/kem-16/HARE-512-kr__keygen.json

Scripts: performance/campaign.py, performance/ngcc_perf.c, performance/hashprof/ in the harness.