ngcc.dev

Performance measurements › arm_1 › kem-17 · system: x86_1 · arm_1

kem-17 Hybrid Equivalent Punctured and Quasi-Cyclic — 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-17/kat.sha256 in the harness) before timing.

instance build flags KAT
hep-qc-1 guide PASS
hep-qc-3 guide PASS
hep-qc-5 guide PASS
hep-qc-7 guide PASS

4. Performance

instance operation mean cycles mean time ops/s median trial n (trials × iters)
hep-qc-1 keygen 213.62 M 79.3 ms 12.6 79.3 ms 100 (5 × 20)
hep-qc-1 enc 11.24 M 4.19 ms 239 4.17 ms 600 (5 × 120)
hep-qc-1 dec 227.93 M 84.6 ms 11.8 84.6 ms 100 (5 × 20)
hep-qc-3 keygen 454.79 M 169 ms 5.93 169 ms 100 (5 × 20)
hep-qc-3 enc 33.33 M 12.4 ms 80.8 12.4 ms 205 (5 × 41)
hep-qc-3 dec 443.20 M 164 ms 6.08 164 ms 100 (5 × 20)
hep-qc-5 keygen 775.55 M 288 ms 3.47 288 ms 100 (5 × 20)
hep-qc-5 enc 75.81 M 28.2 ms 35.5 28.2 ms 100 (5 × 20)
hep-qc-5 dec 751.25 M 279 ms 3.59 276 ms 100 (5 × 20)
hep-qc-7 keygen 3.57 G 1.32 s 0.755 1.32 s 100 (5 × 20)
hep-qc-7 enc 501.64 M 186 ms 5.37 186 ms 100 (5 × 20)
hep-qc-7 dec 3.04 G 1.13 s 0.887 1.13 s 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
hep-qc-1 keygen 325320 1992 KiB 5012 KiB
hep-qc-1 enc 325320 4692 KiB 37524 KiB
hep-qc-1 dec 325320 5164 KiB 11372 KiB
hep-qc-3 keygen 905272 3136 KiB 8804 KiB
hep-qc-3 enc 905272 7052 KiB 43980 KiB
hep-qc-3 dec 905272 8836 KiB 27332 KiB
hep-qc-5 keygen 1896344 1464 KiB 16600 KiB
hep-qc-5 enc 1896344 11052 KiB 52548 KiB
hep-qc-5 dec 1896344 14472 KiB 52180 KiB
hep-qc-7 keygen 12729648 3500 KiB 91248 KiB
hep-qc-7 enc 12729648 54456 KiB 312844 KiB
hep-qc-7 dec 12729648 79036 KiB 338324 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
hep-qc-1 285889 285969 4433 32
hep-qc-3 866210 866298 8978 32
hep-qc-5 1852485 1852581 14421 32
hep-qc-7 12644449 12644577 49297 32

Symmetric primitives

Survey verdict: ICCS-only — never calls the DRNG: zero-initialised PRNG (known kem-17-1/-3)

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
hep-qc-1 keygen 90% 0.0% pseudoXOF 3.67e+04, pseudohash 1
hep-qc-1 enc 55% 0.0% pseudoXOF 14, pseudohash 1, sm3hash 1
hep-qc-1 dec 90% 0.0% pseudoXOF 3.83e+04, pseudohash 1, sm3hash 4
hep-qc-3 keygen 85% 0.0% pseudoXOF 7.32e+04, pseudohash 1
hep-qc-3 enc 54% 0.0% pseudoXOF 14, pseudohash 1, sm3hash 1
hep-qc-3 dec 90% 0.0% pseudoXOF 7.3e+04, pseudohash 1, sm3hash 4
hep-qc-5 keygen 80% 0.0% pseudoXOF 1.17e+05, pseudohash 1
hep-qc-5 enc 50% 0.0% pseudoXOF 14, pseudohash 1, sm3hash 1
hep-qc-5 dec 87% 0.0% pseudoXOF 1.16e+05, pseudohash 1, sm3hash 4
hep-qc-7 keygen 55% 0.0% pseudoXOF 4e+05, pseudohash 1
hep-qc-7 enc 51% 0.0% pseudoXOF 14, pseudohash 1, sm3hash 1
hep-qc-7 dec 77% 0.0% pseudoXOF 3.97e+05, pseudohash 1, sm3hash 4

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
hep-qc-1 KAT log (sha256 6eea8465b3c7593f…) kat/kem-17/hep-qc-1.log
hep-qc-1 timing dec records/kem-17/hep-qc-1__dec.json
hep-qc-1 timing enc records/kem-17/hep-qc-1__enc.json
hep-qc-1 timing keygen records/kem-17/hep-qc-1__keygen.json
hep-qc-1 hash profile dec profile/kem-17/hep-qc-1__dec.json
hep-qc-1 hash profile enc profile/kem-17/hep-qc-1__enc.json
hep-qc-1 hash profile keygen profile/kem-17/hep-qc-1__keygen.json
hep-qc-3 KAT log (sha256 a798e2f658d341fd…) kat/kem-17/hep-qc-3.log
hep-qc-3 timing dec records/kem-17/hep-qc-3__dec.json
hep-qc-3 timing enc records/kem-17/hep-qc-3__enc.json
hep-qc-3 timing keygen records/kem-17/hep-qc-3__keygen.json
hep-qc-3 hash profile dec profile/kem-17/hep-qc-3__dec.json
hep-qc-3 hash profile enc profile/kem-17/hep-qc-3__enc.json
hep-qc-3 hash profile keygen profile/kem-17/hep-qc-3__keygen.json
hep-qc-5 KAT log (sha256 73aad6621f7563f6…) kat/kem-17/hep-qc-5.log
hep-qc-5 timing dec records/kem-17/hep-qc-5__dec.json
hep-qc-5 timing enc records/kem-17/hep-qc-5__enc.json
hep-qc-5 timing keygen records/kem-17/hep-qc-5__keygen.json
hep-qc-5 hash profile dec profile/kem-17/hep-qc-5__dec.json
hep-qc-5 hash profile enc profile/kem-17/hep-qc-5__enc.json
hep-qc-5 hash profile keygen profile/kem-17/hep-qc-5__keygen.json
hep-qc-7 KAT log (sha256 f96dd5e83553d464…) kat/kem-17/hep-qc-7.log
hep-qc-7 timing dec records/kem-17/hep-qc-7__dec.json
hep-qc-7 timing enc records/kem-17/hep-qc-7__enc.json
hep-qc-7 timing keygen records/kem-17/hep-qc-7__keygen.json
hep-qc-7 hash profile dec profile/kem-17/hep-qc-7__dec.json
hep-qc-7 hash profile enc profile/kem-17/hep-qc-7__enc.json
hep-qc-7 hash profile keygen profile/kem-17/hep-qc-7__keygen.json

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