Performance measurements › x86_1 › kem-12 · system: x86_1 · arm_1

kem-12 CTL Algorithm — performance on x86-64 (system x86_1)

Independent measurement following the structure of the NICCS x86 self-assessment guide, §3.5 (1)–(7). Not a submitter self-assessment.

1. Basic information

Category: public-key algorithm; function: key encapsulation
Algorithm: CTL Algorithm
Implementation versions measured: reference
Parameter sets: CTL-257-512, CTL-769-1024, CTL-3329-2048
Security evaluation: kem-12 report
Measurement method: x86_1 method
Submission: NICCS page

2. Assessment environment

item	value
processor	12th Gen Intel(R) Core(TM) i7-12700 (CPU 2, one core)
clock	max 2.10 GHz, governor performance, turbo off, SMT off
memory	31788 MiB
OS / kernel	Debian GNU/Linux 13 (trixie) / 6.12.107+deb13-amd64
compiler / build tool	gcc (Debian 14.2.0-19) 14.2.0 / cmake version 3.31.6
campaign start / end (UTC)	2026-09-25T10:02:21 / 2026-09-28T10:51:21

3. Functional testing (KAT)

Each library was checked against the submitted KAT vectors (SHA-256 manifest kem-12/kat.sha256 in the harness) before timing.

instance	build flags	KAT
`CTL-257-512`	guide	PASS
`CTL-769-1024`	guide	PASS
`CTL-3329-2048`	guide	PASS

4. Performance

instance	operation	mean cycles	mean time	ops/s	median trial	n (trials × iters)
`CTL-257-512`	keygen	17.76 M	8.48 ms	118	8.49 ms	595 (5 × 119)
`CTL-257-512`	enc	61.0 k	29.1 µs	3.43e+04	29.1 µs	100000 (5 × 20000)
`CTL-257-512`	dec	349.0 k	167 µs	6e+03	167 µs	28460 (5 × 5692)
`CTL-769-1024`	keygen	79.80 M	38.2 ms	26.2	38.2 ms	155 (5 × 31)
`CTL-769-1024`	enc	113.7 k	54.3 µs	1.84e+04	54.3 µs	80625 (5 × 16125)
`CTL-769-1024`	dec	711.3 k	340 µs	2.94e+03	340 µs	14345 (5 × 2869)
`CTL-3329-2048`	keygen	2.15 G	1.03 s	0.967	1.04 s	100 (5 × 20)
`CTL-3329-2048`	enc	11.91 M	5.72 ms	175	5.72 ms	880 (5 × 176)
`CTL-3329-2048`	dec	79.73 M	38.4 ms	26	38.4 ms	135 (5 × 27)

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
`CTL-257-512`	keygen	391997	2496 KiB	2940 KiB
`CTL-257-512`	enc	391997	2768 KiB	2944 KiB
`CTL-257-512`	dec	391997	2780 KiB	2952 KiB
`CTL-769-1024`	keygen	392261	2504 KiB	3012 KiB
`CTL-769-1024`	enc	392261	2820 KiB	3000 KiB
`CTL-769-1024`	dec	392261	2888 KiB	3032 KiB
`CTL-3329-2048`	keygen	392317	2528 KiB	6432 KiB
`CTL-3329-2048`	enc	392317	4292 KiB	4580 KiB
`CTL-3329-2048`	dec	392317	4324 KiB	4648 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
`CTL-257-512`	521	2953	473	16
`CTL-769-1024`	1230	6030	1006	32
`CTL-3329-2048`	3009	15617	2353	48

Symmetric primitives

Survey verdict: ICCS-only — sha3.c shim has no permutation and is unused

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
`CTL-257-512`	keygen	8.4%	0.0%	drng 1, pseudoXOF 257
`CTL-257-512`	enc	12%	10%	drng 1, pseudoXOF 3
`CTL-257-512`	dec	6.2%	0.0%	pseudoXOF 4
`CTL-769-1024`	keygen	2.5%	0.0%	drng 1, pseudoXOF 337
`CTL-769-1024`	enc	10%	4.0%	drng 1, pseudoXOF 2, sm3hash 1
`CTL-769-1024`	dec	5.1%	0.0%	pseudoXOF 3, sm3hash 1
`CTL-3329-2048`	keygen	0.1%	0.0%	drng 1, pseudoXOF 260
`CTL-3329-2048`	enc	0.4%	0.1%	drng 1, pseudoXOF 3
`CTL-3329-2048`	dec	0.3%	0.0%	pseudoXOF 4

7. Raw evidence index

Paths are relative to performance/data/x86_1/ in the harness; each JSON file records its commands, environment and trials.

instance	item	file
`CTL-257-512`	KAT log (sha256 `6d4cbf596edd6540…`)	`kat/kem-12/CTL-257-512.log`
`CTL-257-512`	timing dec	`records/kem-12/CTL-257-512__dec.json`
`CTL-257-512`	timing enc	`records/kem-12/CTL-257-512__enc.json`
`CTL-257-512`	timing keygen	`records/kem-12/CTL-257-512__keygen.json`
`CTL-257-512`	hash profile dec	`profile/kem-12/CTL-257-512__dec.json`
`CTL-257-512`	hash profile enc	`profile/kem-12/CTL-257-512__enc.json`
`CTL-257-512`	hash profile keygen	`profile/kem-12/CTL-257-512__keygen.json`
`CTL-769-1024`	KAT log (sha256 `a90c0b008f11c9dd…`)	`kat/kem-12/CTL-769-1024.log`
`CTL-769-1024`	timing dec	`records/kem-12/CTL-769-1024__dec.json`
`CTL-769-1024`	timing enc	`records/kem-12/CTL-769-1024__enc.json`
`CTL-769-1024`	timing keygen	`records/kem-12/CTL-769-1024__keygen.json`
`CTL-769-1024`	hash profile dec	`profile/kem-12/CTL-769-1024__dec.json`
`CTL-769-1024`	hash profile enc	`profile/kem-12/CTL-769-1024__enc.json`
`CTL-769-1024`	hash profile keygen	`profile/kem-12/CTL-769-1024__keygen.json`
`CTL-3329-2048`	KAT log (sha256 `d6c583cccfe71aa0…`)	`kat/kem-12/CTL-3329-2048.log`
`CTL-3329-2048`	timing dec	`records/kem-12/CTL-3329-2048__dec.json`
`CTL-3329-2048`	timing enc	`records/kem-12/CTL-3329-2048__enc.json`
`CTL-3329-2048`	timing keygen	`records/kem-12/CTL-3329-2048__keygen.json`
`CTL-3329-2048`	hash profile dec	`profile/kem-12/CTL-3329-2048__dec.json`
`CTL-3329-2048`	hash profile enc	`profile/kem-12/CTL-3329-2048__enc.json`
`CTL-3329-2048`	hash profile keygen	`profile/kem-12/CTL-3329-2048__keygen.json`

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