Performance measurements › arm_1 › kem-08 · system: x86_1 · arm_1

kem-08 BW-KEM — 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: BW-KEM
Implementation versions measured: reference
Parameter sets: BW_KEM_C128, BW_KEM_C256, BW_KEM_C512
Security evaluation: kem-08 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-08/kat.sha256 in the harness) before timing.

instance	build flags	KAT
`BW_KEM_C128`	guide	PASS
`BW_KEM_C256`	guide	PASS
`BW_KEM_C512`	guide	PASS

4. Performance

instance	operation	mean cycles	mean time	ops/s	median trial	n (trials × iters)
`BW_KEM_C128`	keygen	219.7 k	81.5 µs	1.23e+04	81.1 µs	35480 (5 × 7096)
`BW_KEM_C128`	enc	248.0 k	92 µs	1.09e+04	91.8 µs	33175 (5 × 6635)
`BW_KEM_C128`	dec	286.9 k	106 µs	9.39e+03	106 µs	28990 (5 × 5798)
`BW_KEM_C256`	keygen	559.9 k	208 µs	4.81e+03	208 µs	13730 (5 × 2746)
`BW_KEM_C256`	enc	570.2 k	212 µs	4.73e+03	211 µs	14670 (5 × 2934)
`BW_KEM_C256`	dec	654.4 k	243 µs	4.12e+03	243 µs	12730 (5 × 2546)
`BW_KEM_C512`	keygen	1.94 M	718 µs	1.39e+03	716 µs	4225 (5 × 845)
`BW_KEM_C512`	enc	1.97 M	730 µs	1.37e+03	730 µs	4255 (5 × 851)
`BW_KEM_C512`	dec	2.23 M	828 µs	1.21e+03	827 µs	3810 (5 × 762)

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
`BW_KEM_C128`	keygen	27136	1420 KiB	1492 KiB
`BW_KEM_C128`	enc	27136	1428 KiB	1492 KiB
`BW_KEM_C128`	dec	27136	1428 KiB	1492 KiB
`BW_KEM_C256`	keygen	33472	1428 KiB	1504 KiB
`BW_KEM_C256`	enc	33472	1444 KiB	1508 KiB
`BW_KEM_C256`	dec	33472	1444 KiB	1512 KiB
`BW_KEM_C512`	keygen	38936	1440 KiB	1536 KiB
`BW_KEM_C512`	enc	38936	3516 KiB	3580 KiB
`BW_KEM_C512`	dec	38936	1480 KiB	1544 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
`BW_KEM_C128`	784	1585	768	16
`BW_KEM_C256`	1568	3169	1440	32
`BW_KEM_C512`	3136	6337	2944	64

Symmetric primitives

Survey verdict: ICCS-only — optimized AVX2 auxfunc.c adds an SM3 counter-block fast path

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
`BW_KEM_C128`	keygen	72%	2.0%	drng 1, pseudoXOF 8.02, sm3hash 1
`BW_KEM_C128`	enc	69%	1.8%	drng 1, pseudoXOF 9, sm3hash 1
`BW_KEM_C128`	dec	66%	0.0%	pseudoXOF 10, sm3hash 1
`BW_KEM_C256`	keygen	78%	1.0%	drng 1, pseudoXOF 24.1, pseudohash 1
`BW_KEM_C256`	enc	78%	0.7%	drng 1, pseudoXOF 25, pseudohash 1
`BW_KEM_C256`	dec	72%	0.0%	pseudoXOF 26, pseudohash 1
`BW_KEM_C512`	keygen	83%	0.4%	drng 1, pseudoXOF 24, pseudohash 1
`BW_KEM_C512`	enc	83%	0.3%	drng 1, pseudoXOF 25, pseudohash 1
`BW_KEM_C512`	dec	78%	0.0%	pseudoXOF 26, pseudohash 1

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
`BW_KEM_C128`	KAT log (sha256 `8c140920e969690b…`)	`kat/kem-08/BW_KEM_C128.log`
`BW_KEM_C128`	timing dec	`records/kem-08/BW_KEM_C128__dec.json`
`BW_KEM_C128`	timing enc	`records/kem-08/BW_KEM_C128__enc.json`
`BW_KEM_C128`	timing keygen	`records/kem-08/BW_KEM_C128__keygen.json`
`BW_KEM_C128`	hash profile dec	`profile/kem-08/BW_KEM_C128__dec.json`
`BW_KEM_C128`	hash profile enc	`profile/kem-08/BW_KEM_C128__enc.json`
`BW_KEM_C128`	hash profile keygen	`profile/kem-08/BW_KEM_C128__keygen.json`
`BW_KEM_C256`	KAT log (sha256 `560e3227b48fc50a…`)	`kat/kem-08/BW_KEM_C256.log`
`BW_KEM_C256`	timing dec	`records/kem-08/BW_KEM_C256__dec.json`
`BW_KEM_C256`	timing enc	`records/kem-08/BW_KEM_C256__enc.json`
`BW_KEM_C256`	timing keygen	`records/kem-08/BW_KEM_C256__keygen.json`
`BW_KEM_C256`	hash profile dec	`profile/kem-08/BW_KEM_C256__dec.json`
`BW_KEM_C256`	hash profile enc	`profile/kem-08/BW_KEM_C256__enc.json`
`BW_KEM_C256`	hash profile keygen	`profile/kem-08/BW_KEM_C256__keygen.json`
`BW_KEM_C512`	KAT log (sha256 `d45e123027089069…`)	`kat/kem-08/BW_KEM_C512.log`
`BW_KEM_C512`	timing dec	`records/kem-08/BW_KEM_C512__dec.json`
`BW_KEM_C512`	timing enc	`records/kem-08/BW_KEM_C512__enc.json`
`BW_KEM_C512`	timing keygen	`records/kem-08/BW_KEM_C512__keygen.json`
`BW_KEM_C512`	hash profile dec	`profile/kem-08/BW_KEM_C512__dec.json`
`BW_KEM_C512`	hash profile enc	`profile/kem-08/BW_KEM_C512__enc.json`
`BW_KEM_C512`	hash profile keygen	`profile/kem-08/BW_KEM_C512__keygen.json`

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