Performance measurements › arm_1 › kex-01 · system: x86_1 · arm_1

kex-01 ADKEX (Authenticated Ding Key Exchange) — 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 exchange
Algorithm: ADKEX (Authenticated Ding Key Exchange)
Implementation versions measured: reference
Parameter sets: ADKEX-128, ADKEX-256, ADKEX-512
Security evaluation: kex-01 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 kex-01/kat.sha256 in the harness) before timing.

instance	build flags	KAT
`ADKEX-128`	guide	PASS
`ADKEX-256`	guide	PASS
`ADKEX-512`	guide	PASS

4. Performance

instance	operation	mean cycles	mean time	ops/s	median trial	n (trials × iters)
`ADKEX-128`	exchange	1.33 M	494 µs	2.02e+03	494 µs	6405 (5 × 1281)
`ADKEX-128`	init_a	186	24.4 ns	4.11e+07	24.5 ns	6405 (5 × 1281)
`ADKEX-128`	init_b	175.2 k	65 µs	1.54e+04	64.7 µs	6405 (5 × 1281)
`ADKEX-128`	pass1	372.9 k	138 µs	7.23e+03	138 µs	6405 (5 × 1281)
`ADKEX-128`	pass2	488.5 k	181 µs	5.52e+03	181 µs	6405 (5 × 1281)
`ADKEX-128`	derive_a	290.5 k	108 µs	9.28e+03	108 µs	6405 (5 × 1281)
`ADKEX-128`	derive_b	2845	1.01 µs	9.94e+05	1.01 µs	6405 (5 × 1281)
`ADKEX-256`	exchange	3.64 M	1.35 ms	742	1.35 ms	2285 (5 × 457)
`ADKEX-256`	init_a	184	24.6 ns	4.07e+07	23.7 ns	2285 (5 × 457)
`ADKEX-256`	init_b	532.7 k	198 µs	5.06e+03	198 µs	2285 (5 × 457)
`ADKEX-256`	pass1	1.10 M	407 µs	2.46e+03	405 µs	2285 (5 × 457)
`ADKEX-256`	pass2	1.29 M	477 µs	2.1e+03	476 µs	2285 (5 × 457)
`ADKEX-256`	derive_a	724.0 k	269 µs	3.72e+03	269 µs	2285 (5 × 457)
`ADKEX-256`	derive_b	2828	1 µs	9.99e+05	999 ns	2285 (5 × 457)
`ADKEX-512`	exchange	12.84 M	4.76 ms	210	4.76 ms	650 (5 × 130)
`ADKEX-512`	init_a	191	27.6 ns	3.63e+07	28.8 ns	650 (5 × 130)
`ADKEX-512`	init_b	1.87 M	693 µs	1.44e+03	693 µs	650 (5 × 130)
`ADKEX-512`	pass1	3.84 M	1.43 ms	701	1.42 ms	650 (5 × 130)
`ADKEX-512`	pass2	4.57 M	1.69 ms	590	1.68 ms	650 (5 × 130)
`ADKEX-512`	derive_a	2.60 M	964 µs	1.04e+03	960 µs	650 (5 × 130)
`ADKEX-512`	derive_b	10.6 k	3.91 µs	2.55e+05	3.89 µs	650 (5 × 130)

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
`ADKEX-128`	exchange	30100	1472 KiB	1536 KiB
`ADKEX-128`	init_a	30100	1472 KiB	1536 KiB
`ADKEX-128`	init_b	30100	1472 KiB	1536 KiB
`ADKEX-128`	pass1	30100	1468 KiB	1532 KiB
`ADKEX-128`	pass2	30100	1472 KiB	1536 KiB
`ADKEX-128`	derive_a	30100	1468 KiB	1532 KiB
`ADKEX-128`	derive_b	30100	1472 KiB	1536 KiB
`ADKEX-256`	exchange	30908	3572 KiB	3636 KiB
`ADKEX-256`	init_a	30908	1528 KiB	1592 KiB
`ADKEX-256`	init_b	30908	1528 KiB	1592 KiB
`ADKEX-256`	pass1	30908	3564 KiB	3628 KiB
`ADKEX-256`	pass2	30908	1528 KiB	1592 KiB
`ADKEX-256`	derive_a	30908	3532 KiB	3596 KiB
`ADKEX-256`	derive_b	30908	3556 KiB	3620 KiB
`ADKEX-512`	exchange	32092	3688 KiB	3752 KiB
`ADKEX-512`	init_a	32092	1652 KiB	1728 KiB
`ADKEX-512`	init_b	32092	3640 KiB	3704 KiB
`ADKEX-512`	pass1	32092	3532 KiB	3596 KiB
`ADKEX-512`	pass2	32092	3560 KiB	3624 KiB
`ADKEX-512`	derive_a	32092	1652 KiB	1728 KiB
`ADKEX-512`	derive_b	32092	1652 KiB	1728 KiB

6. Transmission and storage overhead

Bandwidth counts all specified protocol messages and each required public key once. Public keys are transmitted bytes too. Certificates and transport framing are excluded. The published raw timing records are unchanged.

Only the responder's long-term public key is required; this exchange does not authenticate the initiator as a mutually authenticated protocol would.

instance	passes	messages (bytes; raw API)	protocol-message bytes	public key A / B	bandwidth (bytes)	long-term sk (API cap)	shared secret
`ADKEX-128`	2	1600 / 800	2400	0 / 800	3200	1600	32
`ADKEX-256`	2	3168 / 1600	4768	0 / 1568	6336	3136	32
`ADKEX-512`	2	6528 / 3136	9664	0 / 3392	13056	6784	64

Symmetric primitives

Survey verdict: ICCS-only — default DKE_HASH=0; own SM3/HMAC compiled but unreachable; DKE_HASH=2 would switch to SHAKE

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
`ADKEX-128`	exchange	69%	1.2%	drng 3, pseudoXOF 6, sm3hash 563
`ADKEX-256`	exchange	73%	0.4%	drng 3, pseudoXOF 6, sm3hash 1.76e+03
`ADKEX-512`	exchange	84%	0.2%	drng 3, pseudoXOF 10, pseudohash 6, sm3hash 3.7e+03

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
`ADKEX-128`	KAT log (sha256 `c3c89d89e65aa8b9…`)	`kat/kex-01/ADKEX-128.log`
`ADKEX-128`	timing derive_a	`records/kex-01/ADKEX-128__derive_a.json`
`ADKEX-128`	timing derive_b	`records/kex-01/ADKEX-128__derive_b.json`
`ADKEX-128`	timing exchange	`records/kex-01/ADKEX-128__exchange.json`
`ADKEX-128`	timing init_a	`records/kex-01/ADKEX-128__init_a.json`
`ADKEX-128`	timing init_b	`records/kex-01/ADKEX-128__init_b.json`
`ADKEX-128`	timing pass1	`records/kex-01/ADKEX-128__pass1.json`
`ADKEX-128`	timing pass2	`records/kex-01/ADKEX-128__pass2.json`
`ADKEX-128`	hash profile exchange	`profile/kex-01/ADKEX-128__exchange.json`
`ADKEX-256`	KAT log (sha256 `78e3512e964d6163…`)	`kat/kex-01/ADKEX-256.log`
`ADKEX-256`	timing derive_a	`records/kex-01/ADKEX-256__derive_a.json`
`ADKEX-256`	timing derive_b	`records/kex-01/ADKEX-256__derive_b.json`
`ADKEX-256`	timing exchange	`records/kex-01/ADKEX-256__exchange.json`
`ADKEX-256`	timing init_a	`records/kex-01/ADKEX-256__init_a.json`
`ADKEX-256`	timing init_b	`records/kex-01/ADKEX-256__init_b.json`
`ADKEX-256`	timing pass1	`records/kex-01/ADKEX-256__pass1.json`
`ADKEX-256`	timing pass2	`records/kex-01/ADKEX-256__pass2.json`
`ADKEX-256`	hash profile exchange	`profile/kex-01/ADKEX-256__exchange.json`
`ADKEX-512`	KAT log (sha256 `6638a9c454b67760…`)	`kat/kex-01/ADKEX-512.log`
`ADKEX-512`	timing derive_a	`records/kex-01/ADKEX-512__derive_a.json`
`ADKEX-512`	timing derive_b	`records/kex-01/ADKEX-512__derive_b.json`
`ADKEX-512`	timing exchange	`records/kex-01/ADKEX-512__exchange.json`
`ADKEX-512`	timing init_a	`records/kex-01/ADKEX-512__init_a.json`
`ADKEX-512`	timing init_b	`records/kex-01/ADKEX-512__init_b.json`
`ADKEX-512`	timing pass1	`records/kex-01/ADKEX-512__pass1.json`
`ADKEX-512`	timing pass2	`records/kex-01/ADKEX-512__pass2.json`
`ADKEX-512`	hash profile exchange	`profile/kex-01/ADKEX-512__exchange.json`

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