CALL FOR PARTICIPATION

November'96 VERSION of

TOP500 SUPERCOMPUTER LIST


To provide a better basis for statistics on high-performance computers, we have put together the Top500 Supercomputer List. The list is a collection of the sites that have the 500 most powerful computer systems installed.

The next release of our Top500 Supercomputer List will be presented at Supercomputing '96, November 17-22 in Pittsburgh, PA.

The deadline for submitting entries is October 4, 1996.


We are interested in ``NEW ENTRIES'', as well as entries that are no longer valid, on the Top500 List.

Please feel free to contact us at top500@rz.uni-mannheim.de if you have questions.

Thank you for your participation and your cooperation.

Jack J. Dongarra, Hans W. Meuer, Erich Strohmaier


Jack Dongarra <dongarra@cs.utk.edu>
Hans-Werner Meuer <meuer@rz.uni-mannheim.de>
Erich Strohmaier <erich@cs.utk.edu>

Guidelines for Submisson

To provide a better basis for statistics on high-performance computers, we list twice a year the sites that have the 500 most powerful computer systems installed. The best Linpack benchmark performance achieved is used as a performance measure in ranking the computers.

Submissions:

Companies or individuals may submit entries for inclusion to us. Submissions should contain the following informations:



Linpack performance:

Performance numbers for the Linpack benchmark are collected and updated by Jack Dongarra (<dongarra@cs.utk.edu>). Please report any Linpack performance numbers directly to him.

Method of Solution:

In an attempt to obtain uniformity across all computers in performance reporting, the algorithm used in solving the system of equations in the benchmark procedure must confirm to the standard operation count for LU factorization with partial pivoting. In particular, the operation count for the algorithm must be 2/3 n^3 + O(n^2) floating point operations. This excludes the use of a fast matrix multiply algorithm like ``Strassian's Method''. This is done to provide a comparable set of performance numbers across all computers.

Verification:

In addition to cross checking different sources of information, we select randomly a statistical representative sample of the first 500 systems of our database. For these systems we ask the supplier of the information to establish direct contact between the installation site and us to verify the given information. This gives us basic information about the quality of the list in total.

RESTRICTIONS:

As the TOP500 should provide a basis for statistics on the market of high-performance computers, we limit the number of systems installed at vendor sites. This is done for each vendor separately by limiting the accumulated performance of systems at vendor sites to a maximum of 5% of the total accumulated installed performance of this vendor. Rounding is done in favor of the vendor in question.


Back to the main page of the TOP500

top500@rz.uni-mannheim.de
last update: September 1, 1996