Hans Werner Meuer

The marketplace of high-performance computing

In this paper we analyze the major trends and changes in the High-Performance Computing (HPC) market place since the beginning of the journal ‘Parallel Computing’. The initial success of vector computers in the 1970s was driven by raw performance. The introduction of this type of computer systems started the area of ‘Supercomputing’. In the 1980s the availability of standard development environments and of application software packages became more important. Next to performance these factors determined the success of MP vector systems, especially at industrial customers. MPPs became successful in the early 1990s due to their better price/performance ratios, which was made possible by the attack of the ‘killer-micros’. In the lower and medium market segments the MPPs were replaced by microprocessor based symmetrical multiprocessor (SMP) systems in the middle of the 1990s. There success formed the basis for the use of new cluster concepts for very high-end systems. In the last few years only the companies which have entered the emerging markets for massive parallel database servers and financial applications attract enough business volume to be able to support the hardware development for the numerical high-end computing market as well. Success in the traditional floating point intensive engineering applications seems to be no longer suAcient for survival in the market. ” 1999 Elsevier Science B.V. All rights reserved.

The TOP500 Project: Looking Back Over 15 Years of Supercomputing Experience

The TOP500 project was launched in 1993 to provide a reliable basis for tracking and detecting trends in high performance computing. Twice a year, a list of the sites operating the world’s 500 most powerful computer systems is compiled and released.

The TOP500 List and Progress in High-Performance Computing

For more than two decades, the TOP500 list has enjoyed incredible success as a metric for supercomputing performance and as a source of data for identifying technological trends. The projects editors reflect on its usefulness and limitations for guiding large-scale scientific computing into the exascale era.

Changing technologies of HPC

Abstract In 1993 for the first time a list of the top 500 supercomputer sites world-wide was made available. The Top 500 list allows a much more detailed and well-founded analysis of the state of high performance computing. Previously data such as the number and geographical distribution of supercomputer installations were difficult to obtain, and only a few analysts undertook the effort to track the press releases by dozens of vendors. With the Top 500 report now generally and easily available it is possible to present an analysis of the state of High Performance Computing (HPC). This paper summarizes some of the most important observations about HPC as of late 1996, in particular the continued dominance of the world market in HPC by the US, the market penetration by commodity microprocessor based systems, and the growing industrial use of super-computers.

Industrial Application Areas of High-Performance Computing

In 1993, a list of the top 500 supercomputer sites worldwide was made available for the first time. Since then, the Top500 list has been published twice a year. The list allows a detailed and well-founded analysis of the state of high-performance computing (HPC). This article summarizes the recent trends in application areas of HPC systems, focusing on the increase in industrial installations and applications.

The TOP500 Project of the Universities Mannheim and Tennessee

The TOP500 project was initiated in 1993 by Hans Meuer and Erich Strohmaier of the University of Mannheim. The first TOP500 list was published in cooperation with Jack Dongarra,University of Tennessee, at the 8th Supercomputer Conference in Mannheim. The TOP500 list has replaced the Mannheim Supercomputer Statistics published since 1986 and counting the worldwide installed vector systems.

Der Computenik aus Yokohama

Genau zum Stichtag am 15. April 2002 wurde uns für die 19te TOP500 Liste der Earth Simulator (ES) vom Earth Simulator Center in Yokohama gemeldet, mit 35,86 Teraflop/s Linpack Performance klar die neue Nummer Eins. Ehrlich gesagt war das keine Überraschung für mich, schon seit letztem November hatte ich für die 19te Liste den ES an der Spitze erwartet und das auch bei der Vorstellung der 18ten Liste in Denver anlässlich der SC2001 Conference gesagt.*

High Performance Computing in Europe and USA: A Comparison

In the middle of the eighties one author of this paper1started to collect and publish statistics about the supercomputer market. At that time it was rather simple to decide which system qualified as supercomputer. The performance gap between vector system and regular mainframes was too large to leave any doubts. Beginning of the nineties this situation had gradually changed. A considerable number of companies competed in the HPC market with a large variety of architectures such as vector computer, mini vector computer, SIMD (Singel Instruction on Multiple Data) and MPP (Massive Parallel Processing) systems. A new definition was needed to decide which of these systems was a “supercomputer”. This definition needed to be architecture independent. Because of Moore’s Law this definition also had to be dynamic in nature to deal with the constant increase in computer performance.

Changing Technologies of HPC

In the last years the field of High Performance Computing (HPC) faced a major move of their building blocks —the computing nodes— away from proprietary designs towards nodes built out of workstation boards. This movement came along with the success of companies like Silicon Graphics and IBM. As the other companies acting in this field are also moving to CMOS as basic technology these two companies along with Convex/HP are building their HPC systems not only with CPUs but with boards ”off-the-shelf” from successful workstation families. Major signs for the success of this approach are not only the pure number of systems they can sell, but the percentage of systems they are able to sell to industrial users. We will discuss in this paper the different developments based on the Top500 lists of supercomputer sites available since June 1993 [1] and which, for the first time, provide a reliable base for a well-founded analysis of the high-performance computing field. Reports about the situation in 1993, 1994 and 1995 have been published before [2], [3] and [4].

Zuverlässigkeit von Universalrechnern

Fehlverhalten von Computern wird in der Öffentlichkeit manchmal schmunzelnd, zuweilen ärgerlich und meistens mit Schadenfreude zur Kenntnis genommen. So ärgerte sich bekanntlich Bundeskanzler Schmidt vor einiger Zeit über eine vom Computer erstellte und nicht leicht leserliche Wasserrechnung, und schon Anfang der 60er Jahre wurde in der Presse kolportiert, daß eine vom Computer erstellte Rechnung über