 VAMPIRE |
eBACS: ECRYPT Benchmarking of Cryptographic Systems |
 ECRYPT II |
|---|
| General information: | Introduction | eBASH | eBASC | eBAEAD | eBATS | SUPERCOP | XBX | Computers | Arch |
|---|
| How to submit new software: | Tips | hash | stream | aead | dh | kem | encrypt | sign |
|---|
| List of primitives measured: | lwc | sha3 | hash | stream | lwc | caesar | aead | dh | kem | encrypt | sign |
|---|
| Measurements: | lwc | sha3 | hash | stream | lwc | caesar | aead | dh | kem | encrypt | sign |
|---|
| List of subroutines: | verify | decode | encode | sort | core | hashblocks | xof | scalarmult |
|---|
Measurements of hash functions
[Page version: 20250523 10:17:46]
eBASH (ECRYPT Benchmarking of All Submitted Hashes) is a project to measure the performance of hash functions. This page presents benchmark results collected in eBASH:
- Speed (cycles/byte) of hashing an 8-byte message.
- Speed of hashing a 64-byte message.
- Speed of hashing a 576-byte message.
- Speed of hashing a 1536-byte message.
- Speed of hashing a 4096-byte message.
- Speed of hashing a long message.
Graphs show timings for all message lengths between 0 and 4096 bytes. The stair-step shape of a typical curve on the graphs reflects the fact that these hash functions handle messages in blocks, such as 64-byte blocks.
Long-message speed is computed as 1/2048 of the difference in cycle counts between hashing 4096-byte messages and hashing 2048-byte messages, or (starting with supercop-20200409) 1/16384 of the difference in cycle counts between hashing 18208-byte messages and hashing 1824-byte messages. Slightly different computations are used for some hash functions with non-power-of-2 block sizes.
Each table row lists the first quartile of many speed measurements, the median of many speed measurements, the third quartile of many speed measurements, and the name of the primitive. Measurements with large variance are indicated in red with question marks. The symbol T: (starting with supercop-20200816) means that the SUPERCOP database at the time of benchmarking did not list constant time as a goal for this implementation. The symbol T!!! means that constant time was listed as a goal for this implementation, but that the implementation failed TIMECOP. (TIMECOP failures are not necessarily security issues; they can sometimes be resolved by, e.g., declaring that a rejection-sampling condition is safe to declassify.)
There is a separate page with more information about each hash function and each implementation. Designers and implementors interested in submitting new hash functions and new implementations should read the call for submissions.
amd64; Zen 4 (a60f12); 2023 AMD Ryzen 7 7700; 8 x 3800MHz; hertz, supercop-20250415 Measurements Test resultsamd64; Raptor Cove (b06a2); 2023 Intel Core i7-13700H, P cores; 6 x 4800MHz; raptor, supercop-20231107 Measurements Test resultsamd64; Golden Cove (906a4-40); 2022 Intel Core i3-1215U, P cores; 2 x 1600MHz; alder2,1f626960,5600000, supercop-20250415 Measurements Test resultsamd64; Gracemont (906a4-20); 2022 Intel Core i3-1215U, E cores; 4 x 1600MHz; alder2,1f626960,3300000, supercop-20250415 Measurements Test resultsamd64; Golden Cove (90675-00); 2022 Intel Core i3-12100; 4 x 3300MHz; alder, supercop-20250415 Measurements Test resultsamd64; Zen 2 (860f01); 2022 AMD Ryzen 5 4500U; 6 x 3600MHz; renoir, supercop-20250415 Measurements Test resultsamd64; Zen 3 (a50f00); 2021 AMD Ryzen 5 PRO 5650G; 6 x 3900MHz; cezanne, supercop-20250415 Measurements Test resultsamd64; Zen 3 (a50f00); 2021 AMD Ryzen 5 5560U; 6 x 4062MHz; beelink, supercop-20221122 Measurements Test resultsamd64; Zen 2 (860f81); 2021 AMD Ryzen 3 5300U; 4 x 2600MHz; lucienne, supercop-20250415 Measurements Test resultsamd64; Tremont (906c0); 2021 Intel Pentium Silver N6000; 4 x 1100MHz; jasper, supercop-20250415 Measurements Test resultsamd64; Tremont (906c0); 2021 Intel Celeron N5105; 4 x 2000MHz; jasper3, supercop-20250415 Measurements Test resultsamd64; Tremont (906c0); 2021 Intel Celeron N4500; 2 x 1100MHz; jasper2, supercop-20250415 Measurements Test resultsamd64; Zen 3 (a20f10); 2020 AMD Ryzen 5 5600X; 6 x 3700MHz; unstable; spawn, supercop-20231215 Measurements Test resultsamd64; Zen (820f01); 2020 AMD Athlon Silver 3050e; 2 x 1400MHz; dali, supercop-20250415 Measurements Test resultsamd64; Tiger Lake (806c1); 2020 Intel Core i7-1165G7; 4 x 2800MHz; panther, supercop-20250415 Measurements Test resultsaarch64; Cortex-A72 (410fd083); 2019 Broadcom BCM2711; 4 x 1500MHz; pi4b, supercop-20241022 Measurements Test resultsamd64; Zen 2 (830f10); 2019 AMD EPYC 7742; 64 x 2250MHz; rome0, supercop-20250415 Measurements Test resultsamd64; Comet Lake (806ec); 2019 Intel Core i3-10110U; 2 x 2100MHz; comet, supercop-20241022 Measurements Test resultsamd64; Comet Lake (806ec); 2019 Intel Core i3-10110U; 2 x 2100MHz; cubi10, supercop-20250415 Measurements Test resultsamd64; Comet Lake (806ec); 2019 Intel Core i3-10110U; 2 x 2100MHz; know, supercop-20250415 Measurements Test resultsamd64; Ice Lake (706e5); 2019 Intel Core i3-1035G1; 4 x 1000MHz; icelake2, supercop-20221005 Measurements Test resultsamd64; Goldmont Plus (706a8); 2019 Intel Celeron N4020; 2 x 1100MHz; gemini, supercop-20250415 Measurements Test resultsaarch64; Skylark (503f0002); 2018 Ampere eMAG 8180; 32 x 3300MHz; unstable; gcc185, supercop-20240107 Measurements Test resultsaarch64; Cortex-A53 (410fd034); 2018 Broadcom BCM2837B0; 4 x 1400MHz; pi3aplus, supercop-20250415 Measurements Test resultsaarch64; Cortex-A53 (410fd034); 2018 Broadcom BCM2837B0; 4 x 1400MHz; pi3bplus, supercop-20241022 Measurements Test resultsamd64; Coffee Lake (906ea); 2018 Intel Xeon E-2124; 4 x 3300MHz; r24000, supercop-20250415 Measurements Test resultsamd64; Coffee Lake (806ea); 2018 Intel Core i3-8109U; 2 x 3000MHz; like, supercop-20250415 Measurements Test resultsriscv64; U54 (sifive,u54-mc); 2017 SiFive Freedom U540; 4 x 1000MHz; riscvunleashed000, supercop-20240107 Measurements Test resultsamd64; Zen (800f11); 2017 AMD Ryzen 7 1700; 8 x 3000MHz; rumba7, supercop-20250415 Measurements Test resultsamd64; Zen (800f11); 2017 AMD Ryzen 5 1600; 6 x 3200MHz; rumba5, supercop-20241022 Measurements Test resultsamd64; Zen (800f11); 2017 AMD Ryzen 3 1200; 4 x 3100MHz; rumba3, supercop-20221122 Measurements Test resultsamd64; Kaby Lake (906e9); 2017 Intel Xeon E3-1220 v6; 4 x 3000MHz; kizomba, supercop-20250415 Measurements Test resultsamd64; Kaby Lake (806e9); 2017 Intel Core i3-7100; 2 x 2400MHz; whosthere, supercop-20221122 Measurements Test resultsarmeabi; Cortex-A7 (410fc075); 2016 Broadcom BCM2836; 4 x 900MHz; berry2, supercop-20240909 Measurements Test resultsamd64; Goldmont (506c9); 2016 Intel Celeron J3455; 4 x 1500MHz; wooden, supercop-20250415 Measurements Test resultsamd64; Broadwell+AES (406f1); 2016 Intel Xeon E5-2609 v4; 8 x 1700MHz; bolero, supercop-20250415 Measurements Test resultsamd64; Silvermont (406c4); 2016 Intel Atom x5-Z8350; 4 x 1440MHz; cherry, supercop-20250415 Measurements Test resultsamd64; Skylake (506e3); 2015 Intel Xeon E3-1220 v5; 4 x 3000MHz; samba, supercop-20250415 Measurements Test resultsamd64; Airmont (406c3); 2015 Intel Pentium N3700; 4 x 1600MHz; nucnuc, supercop-20241022 Measurements Test resultsamd64; Broadwell+AES (306d4); 2015 Intel Core i3-5005U; 2 x 1900MHz; shoe, supercop-20250415 Measurements Test resultsarmeabi; Cortex-A17 (410fc0d1); 2014 Rockchip RK3288; 4 x 1800MHz; tinker, supercop-20241022 Measurements Test resultsamd64; Haswell+AES (306c3); 2013 Intel Xeon E3-1275 V3; 4 x 3500MHz; titan0, supercop-20250415 Measurements Test resultsamd64; Haswell+AES (306c3); 2013 Intel Xeon E3-1220 v3; 4 x 3100MHz; hiphop, supercop-20231107 Measurements Test resultsamd64; Haswell+AES (306c3); 2013 Intel Core i7-4770; 4 x 3400MHz; speed2supercop, supercop-20250415 Measurements Test resultsamd64; Haswell+AES (306c3); 2013 Intel Core i7-4765T; 4 x 2000MHz; prodesk, supercop-20241022 Measurements Test resultsarmeabi; Cortex-A8 (413fc082); 2012 TI Sitara XAM3359AZCZ100; 1 x 1000MHz; bblack, supercop-20241022 Measurements Test resultsamd64; Bulldozer (600f20); 2012 AMD FX-8350; 4 x 4000MHz; saber214, supercop-20250415 Measurements Test resultsamd64; Ivy Bridge+AES (306a9); 2012 Intel Xeon E3-1275 V2; 4 x 3500MHz; hydra8, supercop-20250415 Measurements Test resultsamd64; Ivy Bridge+AES (306a9); 2012 Intel Core i5-3427U; 2 x 1800MHz; hunsnivy, supercop-20250415 Measurements Test resultsarmeabi; Cortex-A9+NEON (412fc09a); 2011 Freescale i.MX6 Quad; 4 x 1200MHz; unstable; novena, supercop-20220506 Measurements Test resultsarmeabi; ARM1176 (410fb767); 2011 Broadcom BCM2835; 1 x 1000MHz; berry0, supercop-20240909 Measurements Test resultsmipso32; Octeon II (cnmips64v2); 2011 Cavium Octeon II CN6120; 2 x 2000MHz; gcc23, supercop-20230530 Measurements Test resultsamd64; Bobcat (500f10); 2011 AMD G-T56N; 2 x 1650MHz; h8bobcat, supercop-20241022 Measurements Test resultsamd64; K10 32nm (300f10); 2011 AMD A8-3850; 4 x 2900MHz; hydra5, supercop-20250415 Measurements Test resultsamd64; K10 32nm (300f10); 2011 AMD A6-3650; 4 x 2600MHz; hydra4, supercop-20250415 Measurements Test resultsamd64; Bonnell (30661); 2011 Intel Atom D2500; 2 x 1866MHz; h8atom, supercop-20250415 Measurements Test resultsamd64; Sandy Bridge+AES (206a7); 2011 Intel Xeon E3-1225; 4 x 3100MHz; hydra7, supercop-20250415 Measurements Test resultsamd64; Sandy Bridge (206a7); 2011 Intel Core i3-2310M; 2 x 2100MHz; h6sandy, supercop-20241022 Measurements Test resultsamd64; K10 45nm (100f63); 2010 AMD Athlon II Neo K125; 1 x 1700MHz; h3neo, supercop-20250415 Measurements Test resultsamd64; Bonnell (106ca); 2010 Intel Atom N455; 1 x 1000MHz; h2atom, supercop-20241022 Measurements Test resultsamd64; Core 2 45nm (1067a); 2009 Intel Core 2 Duo E7600; 2 x 3060MHz; wolfdale, supercop-20250415 Measurements Test resultsamd64; Core 2 45nm (10676); 2007 Intel Xeon X5450; 8 x 2992MHz; unstable; gcc14, supercop-20220506 Measurements Test resultsamd64; Core 2 65nm (6fb); 2007 Intel Core 2 Quad Q6600; 4 x 2404MHz; margaux, supercop-20250415 Measurements Test resultsamd64; Core 2 65nm (6fb); 2007 Intel Core 2 Duo T7300; 2 x 2000MHz; trident, supercop-20250307 Measurements Test results