AES HDD encryption was XOR

[dan at geer.org]

Victor Duchovni writes:
-+---------------------
 | The computing power of the microprocessor is still under
 | 32 powers of 2 from its inception, naive extrapolation
 | to the next 32 powers of 2 is unwise.

Well taken, indeed.

But what I am myself interested in is the relationship
of the three main up-curves, Moore's for CPU horsepower
per unit of money, and its two un-named siblings for
storage and for bandwidth.  As I read the tea-leaves,
storage is doubling at perhaps a 12-month rate while
bandwidth is faster still.  Yes, these are laboratory
figures, but the lab is where the action is.

This tells me, I think, that the future of computing
is ever more data-rich but, at the same time, that
that data-richness is eclipsed by ever-increasing
data-mobility.

Suppose the doubling times are 18/12/9; then a decade
is two orders of magnitude for CPU, three for storage,
and four for bandwith.  I do not see how this does not
radically alter the economically optimal computing
infrastructure or, for that matter, the nature of the
problems we here are collectively paid to solve.

This is, of course, all irrelevant if and when the
Singularity occurs.  Kurzweil's guess of 2035 is
27 years away, which is to say 18 powers of two out,
not 32.  Perhaps relevant to this list, imagine that
the research described here:

http://technology.newscientist.com/channel/tech/mg20026805.500-cultured-robots-
make-sweet-music-together--.html

was of two programs creating not music but a cipher.

Thinking out loud,

--dan

[ just for amusement, 2008 world production of wheat
  and rice would each cover 53 squares, with maize
  coming in at 51 squares ]

---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo at metzdowd.com