Performance measurements › arm_1 › kem-02 · system: x86_1 · arm_1
kem-02 Amoeba — 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
- Category: public-key algorithm; function: key encapsulation
- Algorithm: Amoeba
- Implementation versions measured: reference
- Parameter sets:
Amoeba128,Amoeba192,Amoeba256,Amoeba384,Amoeba512 - Security evaluation: kem-02 report
- Measurement method: arm_1 method
- Submission: NICCS page
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-02/kat.sha256 in the harness) before timing.
| instance | build flags | KAT |
|---|---|---|
Amoeba128 |
guide | PASS |
Amoeba192 |
guide | PASS |
Amoeba256 |
guide | PASS |
Amoeba384 |
guide | PASS |
Amoeba512 |
guide | PASS |
4. Performance
| instance | operation | mean cycles | mean time | ops/s | median trial | n (trials × iters) |
|---|---|---|---|---|---|---|
Amoeba128 |
keygen | 329.5 k | 122 µs | 8.18e+03 | 122 µs | 24430 (5 × 4886) |
Amoeba128 |
enc | 448.5 k | 166 µs | 6.01e+03 | 166 µs | 18630 (5 × 3726) |
Amoeba128 |
dec | 492.2 k | 183 µs | 5.48e+03 | 182 µs | 16990 (5 × 3398) |
Amoeba192 |
keygen | 448.4 k | 166 µs | 6.01e+03 | 166 µs | 18100 (5 × 3620) |
Amoeba192 |
enc | 622.2 k | 231 µs | 4.33e+03 | 231 µs | 13560 (5 × 2712) |
Amoeba192 |
dec | 689.0 k | 256 µs | 3.91e+03 | 255 µs | 12230 (5 × 2446) |
Amoeba256 |
keygen | 541.3 k | 201 µs | 4.98e+03 | 199 µs | 14235 (5 × 2847) |
Amoeba256 |
enc | 721.2 k | 268 µs | 3.74e+03 | 267 µs | 11700 (5 × 2340) |
Amoeba256 |
dec | 808.7 k | 300 µs | 3.33e+03 | 300 µs | 7865 (5 × 1573) |
Amoeba384 |
keygen | 799.6 k | 297 µs | 3.37e+03 | 295 µs | 9950 (5 × 1990) |
Amoeba384 |
enc | 1.07 M | 398 µs | 2.51e+03 | 397 µs | 7830 (5 × 1566) |
Amoeba384 |
dec | 1.21 M | 447 µs | 2.24e+03 | 447 µs | 6890 (5 × 1378) |
Amoeba512 |
keygen | 1.05 M | 388 µs | 2.58e+03 | 388 µs | 7770 (5 × 1554) |
Amoeba512 |
enc | 1.40 M | 520 µs | 1.92e+03 | 519 µs | 5820 (5 × 1164) |
Amoeba512 |
dec | 1.59 M | 589 µs | 1.7e+03 | 589 µs | 5310 (5 × 1062) |
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 |
|---|---|---|---|---|
Amoeba128 |
keygen | 46780 | 1436 KiB | 1508 KiB |
Amoeba128 |
enc | 46780 | 1448 KiB | 1516 KiB |
Amoeba128 |
dec | 46780 | 1452 KiB | 1520 KiB |
Amoeba192 |
keygen | 46956 | 1440 KiB | 1512 KiB |
Amoeba192 |
enc | 46956 | 1460 KiB | 1524 KiB |
Amoeba192 |
dec | 46956 | 1460 KiB | 1524 KiB |
Amoeba256 |
keygen | 46940 | 1440 KiB | 1520 KiB |
Amoeba256 |
enc | 46940 | 1468 KiB | 1536 KiB |
Amoeba256 |
dec | 46940 | 1468 KiB | 1536 KiB |
Amoeba384 |
keygen | 47100 | 1444 KiB | 1532 KiB |
Amoeba384 |
enc | 47100 | 1484 KiB | 1552 KiB |
Amoeba384 |
dec | 47100 | 1488 KiB | 1556 KiB |
Amoeba512 |
keygen | 47148 | 1448 KiB | 1544 KiB |
Amoeba512 |
enc | 47148 | 1504 KiB | 1572 KiB |
Amoeba512 |
dec | 47148 | 1508 KiB | 1576 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 |
|---|---|---|---|---|
Amoeba128 |
784 | 1712 | 1047 | 64 |
Amoeba192 |
1144 | 2504 | 1581 | 64 |
Amoeba256 |
1648 | 3440 | 1833 | 64 |
Amoeba384 |
2440 | 5096 | 2769 | 64 |
Amoeba512 |
3520 | 7040 | 3914 | 64 |
Symmetric primitives
Survey verdict: ICCS-only
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 |
|---|---|---|---|---|
Amoeba128 |
keygen | 81% | 3.4% | drng 2, pseudoXOF 20.5 |
Amoeba128 |
enc | 79% | 1.2% | drng 1, pseudoXOF 23, sm3hash 4 |
Amoeba128 |
dec | 73% | 0.0% | pseudoXOF 23, sm3hash 4 |
Amoeba192 |
keygen | 80% | 2.5% | drng 2, pseudoXOF 23.5 |
Amoeba192 |
enc | 79% | 0.9% | drng 1, pseudoXOF 26, sm3hash 4 |
Amoeba192 |
dec | 71% | 0.0% | pseudoXOF 26, sm3hash 4 |
Amoeba256 |
keygen | 79% | 2.0% | drng 2, pseudoXOF 26.5 |
Amoeba256 |
enc | 76% | 0.8% | drng 1, pseudoXOF 28, sm3hash 4 |
Amoeba256 |
dec | 68% | 0.0% | pseudoXOF 28, sm3hash 4 |
Amoeba384 |
keygen | 79% | 1.4% | drng 2, pseudoXOF 36.5 |
Amoeba384 |
enc | 76% | 0.5% | drng 1, pseudoXOF 38, sm3hash 4 |
Amoeba384 |
dec | 67% | 0.0% | pseudoXOF 38, sm3hash 4 |
Amoeba512 |
keygen | 78% | 1.1% | drng 2, pseudoXOF 46.6 |
Amoeba512 |
enc | 75% | 0.4% | drng 1, pseudoXOF 47, sm3hash 4 |
Amoeba512 |
dec | 66% | 0.0% | pseudoXOF 47, 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 |
|---|---|---|
Amoeba128 |
KAT log (sha256 c069c1acff6568af…) |
kat/kem-02/Amoeba128.log |
Amoeba128 |
timing dec | records/kem-02/Amoeba128__dec.json |
Amoeba128 |
timing enc | records/kem-02/Amoeba128__enc.json |
Amoeba128 |
timing keygen | records/kem-02/Amoeba128__keygen.json |
Amoeba128 |
hash profile dec | profile/kem-02/Amoeba128__dec.json |
Amoeba128 |
hash profile enc | profile/kem-02/Amoeba128__enc.json |
Amoeba128 |
hash profile keygen | profile/kem-02/Amoeba128__keygen.json |
Amoeba192 |
KAT log (sha256 ac9ab896ac650fc9…) |
kat/kem-02/Amoeba192.log |
Amoeba192 |
timing dec | records/kem-02/Amoeba192__dec.json |
Amoeba192 |
timing enc | records/kem-02/Amoeba192__enc.json |
Amoeba192 |
timing keygen | records/kem-02/Amoeba192__keygen.json |
Amoeba192 |
hash profile dec | profile/kem-02/Amoeba192__dec.json |
Amoeba192 |
hash profile enc | profile/kem-02/Amoeba192__enc.json |
Amoeba192 |
hash profile keygen | profile/kem-02/Amoeba192__keygen.json |
Amoeba256 |
KAT log (sha256 11335a6ea68b71c8…) |
kat/kem-02/Amoeba256.log |
Amoeba256 |
timing dec | records/kem-02/Amoeba256__dec.json |
Amoeba256 |
timing enc | records/kem-02/Amoeba256__enc.json |
Amoeba256 |
timing keygen | records/kem-02/Amoeba256__keygen.json |
Amoeba256 |
hash profile dec | profile/kem-02/Amoeba256__dec.json |
Amoeba256 |
hash profile enc | profile/kem-02/Amoeba256__enc.json |
Amoeba256 |
hash profile keygen | profile/kem-02/Amoeba256__keygen.json |
Amoeba384 |
KAT log (sha256 e13fb3e66bab01a0…) |
kat/kem-02/Amoeba384.log |
Amoeba384 |
timing dec | records/kem-02/Amoeba384__dec.json |
Amoeba384 |
timing enc | records/kem-02/Amoeba384__enc.json |
Amoeba384 |
timing keygen | records/kem-02/Amoeba384__keygen.json |
Amoeba384 |
hash profile dec | profile/kem-02/Amoeba384__dec.json |
Amoeba384 |
hash profile enc | profile/kem-02/Amoeba384__enc.json |
Amoeba384 |
hash profile keygen | profile/kem-02/Amoeba384__keygen.json |
Amoeba512 |
KAT log (sha256 4aeac966c6f66723…) |
kat/kem-02/Amoeba512.log |
Amoeba512 |
timing dec | records/kem-02/Amoeba512__dec.json |
Amoeba512 |
timing enc | records/kem-02/Amoeba512__enc.json |
Amoeba512 |
timing keygen | records/kem-02/Amoeba512__keygen.json |
Amoeba512 |
hash profile dec | profile/kem-02/Amoeba512__dec.json |
Amoeba512 |
hash profile enc | profile/kem-02/Amoeba512__enc.json |
Amoeba512 |
hash profile keygen | profile/kem-02/Amoeba512__keygen.json |
Scripts: performance/campaign.py, performance/ngcc_perf.c, performance/hashprof/ in the harness.