[SATLUG] question of the measurement of processing power past and
Enrique Sanchez Vela
esanchezvela at yahoo.com
Tue Aug 12 17:13:44 CDT 2008
--- On Tue, 8/12/08, Todd W. Bucy <toddwbucy at grandecom.net> wrote:
> From: Todd W. Bucy <toddwbucy at grandecom.net>
> Subject: [SATLUG] question of the measurement of processing power past and present
> second question
> is it generally true that the ability to code software (at
> a theoretical
> level at least) has always outstripped the ability of the
> hardware to
> process such software? In other words software developers
> in general
> have the ability to write code which demands more from the
> hardware (in terms of resources/computing power) then what
> the hardware
> is capable of doing?
> thanks for any help
that question has many angles, for example a badly structured SQL query can bring a system to its knees as well a missing index, but it is also true that there is a physical limit on what machines can do.
in the other hand, there are problems that have not been tackled or there are disregarded variables to make problems less complicated and solvable with the available or procured hardware, now people are asking newer questions or want to include otherwise disregarded variables on real time simulations.
I am willing to bet that if you bring a machine 10 times faster than RoadRunner (http://www.top500.org/system/9485) researchers will bring a problem big enough to keep it running for days.
More information about the SATLUG