In comparison to 10 years ago the processor scene has become drastically different. While in the period 1980--1990, the proprietary processors and in particular the vectorprocessors were the driving forces of the supercomputers of that period, today that role has been taken on by common off-the-shelf RISC processors and Intel x86 compatible processors. In fact there are only two companies left that produce vector systems while all other systems that are offered are based on RISC CPUs or x86-like (except the Cray MTA-2). Therefore it is useful to give a brief description of the main processors that populate the present supercomputers and look a little ahead to the processors that will follow in the coming year.
The RISC processor scene has shrunken drastically in the last year. Although the HP/Compaq EV7 and the HP PA-RISC processors are still produced, systems containing the former chip will only be marketed until 2006 due to HP's decision to replace these processors by the Itanium line of processors. The PA-RISC 8800 is still made for the Superdome 9000 machine but will also disappear shortly, although officially a PA-RISC 8900 still will be produced.
Although at this moment the RISC processors still play a role many of these will disappear from the scene: the withdrawal of systems containing the Alpha, PA-RISC and MIPS processors is announced all in favour of the Itanium processor product line. The disappearance of RISC processor families demonstrates a trend that is both worrying and interesting: worrying because the diversity in the processor field is decreasing severely and, with it, the choice for systems in this sector. On the other hand there is the trend to enhance systems having run-of-the-mill processors with special-purpose add-on processors in the form of preconfigured FPGAs or DSPs because their possibilities in performance, price level, and ease of use has improved to a degree that they offer attractive alternatives for certain application fields.
The RISC processors generally have a clock frequency that is lower than that of
the Intel Pentium 3/4 processors or the corresponding AMD Intel look-alikes.
However, they have a number of facilities that put them ahead in the speed of
floating-point oriented applications. Firstly, all RISC processors are able to
deliver 2 or more 64-bit floating-point results in one clock cycle. Secondly,
all of them feature out-of-order instruction execution, which enhances the
number of instructions per cycle that can be processed (although the newer AMD
processors also have 2-way floating-point instruction issuing and out-of-order
execution, they are somewhat limited by their adherence to the Intel x86
instruction set). The bandwidth from the processor to the memory, in case of a
cache miss, used to be larger than that of the Intel(-like) processors.
For the AMD Opteron, however, this is not the case anymore, because of its
on-chip memory controller and HyperTransport bus from memory to CPU.
Notwithstanding the commonalities between the various RISC processors, there
are also differences in instruction latencies, number of instructions
processed, etc., which we will address below. We provide block diagrams for
each of the processors to give a schematic idea of their structure. However,
these figures do not reflect the actual layout of the devices on the respective
chips.