Author: Sune Fischer
Date: 02:47:25 11/13/01
Go up one level in this thread
On November 12, 2001 at 18:57:25, Vincent Diepeveen wrote: >Working with floating point is dead slow on PC's. The whole top500 list >is partly a big joke because most are clusters from which not even >in theoretic cases it's possible to use all nodes at the same time >for a single program! > >Not to mention the fact that for nearly any application on earth, clusters >are slower than the same number of cpu's in a shared memory environment. > >Best regards, >Vincent They use the LINPACK to benchmark them for the GFlops: "The benchmark used in the LINPACK Benchmark is to solve a dense system of linear equations. For the TOP500, we used that version of the benchmark that allows the user to scale the size of the problem and to optimize the software in order to achieve the best performance for a given machine. This performance does not reflect the overall performance of a given system, as no single number ever can. It does, however, reflect the performance of a dedicated system for solving a dense system of linear equations. Since the problem is very regular, the performance achieved is quite high, and the performance numbers give a good correction of peak performance." http://www.top500.org/lists/linpack.html But the list is not entirely correct for other reasons. I know of a few machines that should have made it to the list around the 250 spot. It was a few years ago, but never they appeared on it. It is a shame, because what is the point if it isn't _the_ top 500. -S.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.