From: Perry E.Metzger (perry@piermont.com)
Date: Wed Nov 26 2003 - 17:11:11 MST
James Rogers <jamesr@best.com> writes:
> The amount of parallelism we can manage depends greatly on the nature of
> the parallelism. Our hardware doesn't do it because our algorithm
> models have limited use for it. Again, most of our parallelism is the
> low ROI kind, which gives limited economic incentive to bother with it.
Actually, most CPUs sold today do SIMD of various kinds for doing
multimedia work. There are a lot of easy parallel algorithms in things
like graphics, and modern processors take advantage of that. Modern
supercomputers tend to be massively parallel, too. All weather
prediction is done on such machines. Furthermore, even standard
computation is done these days by pipelining techniques that
essentially turn single processors into mildly (say, eight way)
parallel devices.
The issue is the use of highly non-Von Neumann like architectures for
every day use. Generally speaking, we haven't found comfortable
programming models or applications for such beasts.
-- Perry E. Metzger perry@piermont.com
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:43 MDT