aboutsummaryrefslogblamecommitdiffstats
path: root/README
blob: f2abbc1b4422c6c9453fc88e637f298d2dcf0086 (plain) (tree)













































































                                                                     









                                                                     
Checksum utility for Keccak and its close
derivatives SHA-3, SHAKE and RawSHAKE.

---------------------------------------------------------------------

Performance benchmark on a 1 GiB file (yes,
that is quite large) with /dev/urandom data:

$ keccak-224sum
real	0m9.550s
user	0m9.223s
sys	0m0.320s

$ keccak-256sum
real	0m9.841s
user	0m9.560s
sys	0m0.260s

$ keccak-384sum
real	0m12.902s
user	0m12.580s
sys	0m0.313s

$ keccak-512sum
real	0m18.084s
user	0m17.823s
sys	0m0.243s

$ keccaksum
real	0m10.438s
user	0m10.140s
sys	0m0.287s

$ sha1sum
real	0m3.421s
user	0m3.157s
sys	0m0.253s

$ sha224sum
real	0m8.196s
user	0m7.910s
sys	0m0.263s

$ sha256sum
real	0m9.361s
user	0m9.097s
sys	0m0.257s

$ sha384sum
real	0m5.450s
user	0m5.223s
sys	0m0.220s

$ sha512sum
real	0m4.507s
user	0m4.247s
sys	0m0.257s

$ md5sum
real	0m2.987s
user	0m2.703s
sys	0m0.283s

$ md6sum
real	1m28.873s
user	1m28.470s
sys	0m0.360s


Note:
The times are a bit random, and only one run
has been made. "user" is the most interesting
data. Additionally, only the pure keccak variants
have been tested because the difference is
between them and the other are quite small;
and of course the other hashing utilities from
other packages using other algorithm-families.

---------------------------------------------------------------------

The implementation of the algorithms is done
in libkeccak (https://github.com/maandree/libkeccak),
so to verify that this program is correct,
verify libkeccak using its test suite
(https://github.com/maandree/libkeccak/blob/master/src/test.c)
and check that some of hashes are returned identically
by sha3sum.