This is the eighth edition of a report in which we attempt to give an overview of parallel and vector systems that are commercially available or are expected to become available within a short time frame (typically a few months to half a year). We choose the expression ``attempt'' deliberately because the market of parallel- and vector machines is highly evasive: the rate with which systems are introduced --- and disappear again --- is very high and therefore the information will probably be only approximately valid. Nevertheless, we think that such an overview is useful for those who want to obtain a general idea about the various means by which these systems strive at high performance, especially when it is updated on a regular basis.
We will try to be as up-to-date and compact as possible and on these grounds we think there is a place for this report. Like last year, the present report will be somewhat shorter than earlier ones: at this moment systems are disappearing at a faster rate than new ones replace them. The reasons for this seem to be threefold:
These effects make the \hp computing scene somewhat more clear (and also somewhat less adventurous). Still, the supercomputer market is very dynamic and we cannot hope to give a complete report for the reason already mentioned above: the speed with which companies and systems appear and disappear makes this almost impossible. However, by updating the report we can at least follow the main trends in popular and emerging architectures.
The rules for including systems is as follows: they should be either available commercially at the time of appearance of this report, or within 6 months thereafter. This excludes interesting resarch systems like the Intel ASCI Option Red system at Sandia National Laboratory (with a measured performance of 1.3 Tflop/s), the CP-PACS at the University of Tsukuba (measured performance of 368 Gflop/s) and the Numerical Wind Tunnel at the National Aerospace Lab. in Japan (230 Gflop/s), because they are not marketed and only available at the institutes mentioned and, therefore, of not much benefit to the supercomputer community at large.
The rule that systems should be available within a timespan of 6 months
is to avoid confusion by describing systems that are announced much too
early, just for marketing reasons and that will not be available to
general users within a reasonable time. We also have to refrain from
including all generations of a system that are still in use. Therefore,
for instance, we do not include the Convex C4600 series or the Cray C90
series, or the Thinking Machines CM-5 anymore although these systems
are still used widely. Generally speaking, we include machines that
are still marketed or will be marketed within 6 months. To add to
the information given in this report, we quote the Web addresses of the
vendors because the information found there may be more recent than
what can be provided here. On the other hand, such pages should be read
with care because it will not always be clear what the status is of the
products described there.
Some vendors offer systems that are identical in all respects except in
the clock cycle of the nodes (examples are the Digital Alphaserver and
the Fujitsu AP3000). In these cases we always only mention the models
with the fastest clock as it will be always possible to get the slower
systems and we presume that the reader is primarily interested in the
highest possible speeds that can be reached with these systems.
The availability of relatively low-cost (RISC) processors and network products to connect these processors together with standardised communication software have with has stimulated the building of home-brew clusters computers as an alternative to complete systems that offered by vendors. Obviously, we cannot list all the variants that are and have been built by groups all over the world.
We order the systems by their various architectural classes, which should facilitate to find the information of systems that belong to a certain class. We also omit the price information which in most cases is next to useless. If available, we will give some information about performances of systems based on user experiences instead of only giving theoretical peak performances. Here we have adhered to the following policy: We try to quote best measured performances, if available, thus providing a more realistic upper bound than the theoretical peak performance. We hardly have to say that the speed range of supercomputers is enormous, so the best measured performance will not always reflect the performance of the reader's favourite application. When we give performance information, it is not always possible to quote all sources and in any case if this information seems (or is) biased, this is entirely the responsibility of the author of this report. He is quite willing to be corrected or to receive additional information from anyone who is in the position to do so.
Before giving a recount of the systems proper, we first define the architectural classes and some other terms in the section on architecture which will be used in the section that gives a recount of the (almost) available machines. There is also a section that lists the some systems have disappeared from the market and a section that presents some systems that are under development and have a fair chance to appear on the market.
The overview given in this report concentrates on the computational capabilities of the systems discussed. To do full justice to all assets of present days high-performance computers one should list their I/O performance and their connectivity possibilities as well. However, the possible permutations of configurations even for one model of a certain system often are so large that they would multiply the volume of this report, which we tried to limit for greater clarity. So, not all features of the systems discussed will be present. Still we think (and certainly hope) that the impressions obtained from the entries of the individual machines may be useful to many. We also omitted some systems that may be characterised as "high-performance" in the fields of database management, real-time computing, or visualisation. Therefore, as we try to give an overview for the area of general scientific and technical computing, systems that are primarily meant for database retrieval like the AT\&T GIS systems or concentrate exclusively on the real-time user community, like Concurrent Computing Systems, are not discussed in this report.
Although most terms will be familiar to many readers, we still think it is worthwhile to give some of the definitions in the achictecture section because some authors tend to give them a meaning that may slightly differ from the idea the reader already has acquired.
Lastly, we should point out that the WWW version is available at
various places. The URLs are:
USA:
www.netlib.org/utk/papers/advanced-computers/paper.html
Europe:
www.nwo.nl/ncf/indexeng.htm under the "Available
Documentation" button.
Europe:
www.fys.ruu.nl/~steen/overview/overview98.html.