7 Conclusions

From the present six releases of the TOP500 we see:

• For positions in the range of 100--500 the performance of the individual systems is increasing by a factor of 2 every year while the total installed performance is increasing by a factor of 1.8 every year.
• First systems of the next generation of HPC systems are showing up in the list. From this you can expect a new system in position one next year.
• Japan has no SMP systems installed till now.
• The US and Japanese vendors are dominating their home markets, while European manufacturers are playing only a minor role even in their own region.
• SGI and IBM are leading the list with respect to the number of installed systems, while Cray Research is still in front with respect to installed performance.
• MPP systems are the dominating architecture, while the number of PVP systems is steadily going down in the TOP500.
• SMP workstation systems have reached a very impressive portion of the systems in the TOP500, but seem to have reached their climax, while at the same time cluster of SMPs are entering the list for the first time.
• The number of ECL based systems is strongly decreasing all the time, and by the end of 1995 about 80% of the systems in the TOP500 were built with CMOS technology.
• In the TOP500 a strong trend to nodes being binary-compatible to major workstation families can be seen.
• Vendors using such "off-the-shelf" nodes (IBM, SGI and Convex) are in the position to sell many systems to industrial customers overproportionally.

With the TOP500 project going into its fourth year, many trends and evolutions of the HPC market could be made quite transparent. This has proven the TOP500 to be a very valuable tool. Some of the trends mentioned can surely be stated and anticipated without the TOP500 while many others are certainly surprising and could not be visualized without it.