Herb fought the law--Amdahl's Law, that is--and Herb won

Herb fought the law--Amdahl's Law, that is--and Herb won

Post by Amal » Sun, 17 Feb 2008 07:42:59


Hi,

This is in reference to Herb Sutter's article in DDJ. Link for the
article is http://www.yqcomputer.com/

The subtitle of the article is "Herb fought the law--Amdahl's Law,
that is--and Herb won". It does make a feel like herb did something to
make Amdahl's law stand behind. If someone has ever tried to break the
Amdahl's law it is Gustafson. And even Gustafson named his paper
"Reevaluating Amdahl's Law"

See the article http://www.yqcomputer.com/

Thanks and regards,
Amal P

--
 
 
 

Herb fought the law--Amdahl's Law, that is--and Herb won

Post by eugen » Sun, 24 Feb 2008 01:48:37

In article < XXXX@XXXXX.COM >,


The first question would be Who is Herb?" So it comes up with the one
liner at MS and C++. Neither of which have distinguished themselves in
high (whatever: performance, end, etc.) computing. I don't see how Herb
won because he only talks about O(n) and O(k), not O(n^2), nor O(n^3),
nor higher orders of n which is pretty common much less P and NP. MS is
what you get for commodity. I can't think of anyone more deserving of
those orders than nuclear bomb designers (I was informed that Gates just
visited LLNL). Amdahl sleeps pretty well at night.


Amdahl sleeps pretty well at night.

The world needs lots of instances not isolated special cases.

--