[Cryptography] Is there a good algorithm providing both compression and encryption at the same time?

Tom Mitchell mitch at niftyegg.com
Fri May 8 01:20:45 EDT 2015


On Wed, May 6, 2015 at 1:15 AM, Francois BERENGER <
francois.berenger.fun at gmail.com> wrote:

> Dear list,
>
> While programming an open source distributed system, I discovered
> it could be interesting for the system to use a well established algorithm
> that can do these two things at the same time (for performance reasons).
>

With multi core processors piping the output of a compressor
in to an encryption tool (unix style) would get two cores active.
This would solve much of the performance issues.
Attempting to write a single secure pile of code to do it all would
be difficult to get correct.

On modern windows systems there is commonly enough RAM for buffering
the output of one stage writing to a temp  file. If so this buffering has
little impact
even on a single core machine.  IPC via local shared memory pipes or via
physically
secure network links could also prove valuable tricks to improve
performance while keeping
the individual code bases almost untouched.


> I was thinking about something along those lines:
>
> compression(clear_text) = (compression_dictionary, compressed_text)
>

Compression dictionaries for some specific content will take you back
to the old time dictionaries
http://en.wikipedia.org/wiki/Commercial_code_%28communications%29
Some of these dictionaries were guarded as company secrets i.e. a code book.
Computing a file by file dictionary based compression could be partitioned
for parallel computation and might have enough speedup to have value.

Some compilers have flags that disclose optimization wins and losses.
Some of these hints can be used to improve the code base in ways
that the compiler can optimize things.

-- 
  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/20150507/fc552f87/attachment.html>


More information about the cryptography mailing list