Introduction

- In 1994 high performance computing in the U.S. appears to be undergoing a major transition. With the introduction of powerful new massively parallel processors (MPP) by established vendors in the high performance computing field such as the T3D by Cray Research, the SP-2 by IBM, and the Exemplar SPP by Convex there are strong indications that MPP is now finally in the mainstream of scientific supercomputing. These new entries compete with existing MPP vendors such as Intel, Kendall Square, Ncube, and Thinking Machines for shares of a market, which is not as rapidly increasing as expected. Traditional parallel vector supercomputes such as the Cray C90 continue to dominate both performance benchmarks [1], and show strong results when the breadth and complexity of applications is concerned (see [9]). The strength of the market (more in terms of expectations, than in real dollars) can be seen from the list of vendors in Table 10.1, which is updated from Smaby [11]. The consequences of such a large number of vendors competing for such a small (but highly visible and important) market are widely discussed, for example in [10].

[table183]

At the same time the federal High Performance Computing and Communications Program (HPCC) is in high gear and considerable progress has been made as documented in the famous `Blue Book' [3]. On the other hand there has been criticism on the this very program, e.g., in the form of a report by the Congressional Budget Office [4]. The discussion about HPC in the commercial and in the government market place is mainly based on beliefs and impressions, and often lacks hard data. It is surprising that a field such as HPC that is deemed so critically important to the national agenda lacks almost completely any quantitative assessment of its progress.

In this report it is attempted to shed some light on recent developments in HPC in the U.S. and present some quantitative data on the type and distribution of HPC technology. All the information here is based on the TOP500 list of November 1993. The report [7] ranks the 500 top performing supercomputers worldwide. The measure of performance is the maximal achieved [tex2html_wrap1326] value for the computer on the LINPACK benchmark as reported in [5]. Using this measure the cutoff to make the list of the TOP500 systems worldwide is a performance of 472 Mflop/s on the LINPACK benchmark. Six Fujitsu VP-200E computers take on the ranks from 495 to 500. Interestingly the top ranked machine is also made by Fujitsu: the 140 processor specially built computer for the Numerical Windtunnel project at NAL in Japan is rated at 124.5 Gflop/s.

Before investigating some of the data in [7] in more detail, it is important to understand the limitations of the TOP500 study. These limitations can be summarized as follows:

• The list is a snapshot in time. The TOP500 list reflects the state of supercomputer installations as of November 1993. Since the market is extremely dynamic, such a study is almost out of date, by the moment it is published. For example the November 1993 report does not list any of the Cray T3D installations, by March 94 there are about 15. It lists only six IBM SP-1 machines, and by March 94 there are more than 75 installed. However, since the data were collected consistently, we have at least an accurate reflection of the distribution at this time.

• The LINPACK [tex2html_wrap1328] value overestimates the performance of parallel machines. It is known that solving dense linear systems is a task that is well suited to highly parallel systems. MPPs therefore tend to exhibit a proportionally much higher performance than parallel vector processors (PVP). Even though the NAS Parallel Benchmarks [2], for example, would be a more accurate measure of parallel system performance, LINPACK figures are more widely available. They also provide an upper limit of the best possible performance of a computer on any application. If understood this way, the results in TOP500 take on a different, but at least consistent meaning.

• A large number of systems are in classified or highly proprietary areas. Both government and industrial users of HPC are for different reasons reluctant to provide details on the type and number of systems used. This may lead to double counts in some cases, and to undercounts in others, for example a large Ncube2 system to be rumored to be installed at Shell is not on the list. On the other hand the TOP500 contains many `government' listings where location and use is competely unknown, and where is it possible that the same machine is counted twice.

• The TOP500 does not reflect actual usage of machines. For example several of the Intel Paragon machines are listed with high [tex2html_wrap1330] value, however, it is known that many of the larger Paragon systems in 1993 were not yet able to deliver their full compute power for application users (see for example [12] for machine #48). Other systems, such as the Cray 2 at Phillips Laboratory (#235) have been decommissioned and are no longer in active use. However, there is no reasonable way to consistently estimate the actual compute cycles delivered by any of the computers on the list.

In spite of these inherent limitations the TOP500 can provide extremely useful information, and valuable insights. It is more accurate than many marketing studies, and the possible sources of error discussed above are probably statistically insignificant, if we consider only summary statistics, and not individual data. All Mflop/s or Gflop/s performance figures here refer to performance in terms of Linpack [tex2html_wrap1332].

In the analysis of geographical distribution, machines in Canada have been included in the figures for the U.S., and the figures for Europe include all European countries, not just EC members. The other country category includes mostly countries of the Pacific Rim with the exclusion of Japan, and a few Latin American Countries.