[Cryptography] Creating a Parallelizeable Cryptographic Hash Function

Tom Mitchell mitch at niftyegg.com
Fri Oct 3 20:19:45 EDT 2014


On Fri, Oct 3, 2014 at 8:15 AM, Jerry Leichter <leichter at lrw.com> wrote:

> On Oct 2, 2014, at 10:37 PM, Sandy Harris <sandyinchina at gmail.com> wrote:
> > There has been a lot of work on parallelizable hashing. Web search for
> > "tree hashing" will turn up much of it.
>
......

> Keep in mind that "parallelizable" is often taken to mean "linear in the
> number of available processors".  No tree algorithm is "parallelizable" in
> this sense - it has a logarithmic delay to roll up the results.
>

Minor point that should not be ignored.

To a programmer a good hash table is not the same as a good crypto hash.
A programmer simply wants a fast lookup with a minimum miss, collision.
Most programmers do not care if a collision is moderately easy to  fabricate
because they want to get close enough not exactly and will walk their way to
the desired data (short walk).

Crypto hashes need to be nearly impossible to generate by altering the data
input and spoofing a match.

Thus a fast hash for a Google webpage lookup is not the same design need as
a fast hash
for Google data that should be kept secret and private.

-- 
  T o m    M i t c h e l l
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20141003/6a5c5734/attachment.html>


More information about the cryptography mailing list