Parallel Processing
and Super-Scalar
Through this chapter, the reader is introduced to some basic concepts of parallel process-
ing using multi-processors and also to the super-scalar operation. Incidentally, both are
generally adopted for larger and heavy-duty computers, known as super computers. After
completion of this chapter, the reader would know about
R Some fundamental features of parallel processing including memory handling.
R Different types of network topologies and their applicability.
R Super-scalar operation and processors.
So far, we have limited our discussions within those computers that are controlled by single pro-
cessor. As we have already discussed in Chapter 1, the scope of these machines are limited to the
usage as a personal computer or similar design and research activities. In this context we have also
discussed about multitasking and pipeline architecture which enhance the throughput of a one-
processor system.
However, there are some specialized areas where more powerful machines are demanded. For exam-
ple, the three-dimensional mapping of the galaxy, several areas of quantum mechanics, nite-element
method of structural analysis are only a few of such areas where our personal computers would not
be able to handle the volume of the large quantity of data sets or produce accurate results within the
stipulated timeframe. In these areas, larger and more powerful computers are required which are able
to perform larger volume of arithmetical operations at a very fast rate than our normal PCs. In general,
these computers are equipped with more than one processor and designed around the architecture which
is familiar as parallel processing.
M13_GHOS1557_01_SE_C13.indd 391M13_GHOS1557_01_SE_C13.indd 391 4/29/11 5:25 PM4/29/11 5:25 PM

Get Computer Architecture and Organization now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.