Re:Nuts and Bolts of Multithreaded Programming (2011)
by nine_k on Sunday July 02, 2017 @06:53PM
(#14684739)
Can a SIMD system be really considered "multi-threaded"?
[ Parent
| on HN ]
Re:Nuts and Bolts of Multithreaded Programming (2011)
by chrisseaton on Sunday July 02, 2017 @08:28PM
(#14685070)
SIMD is the classic example of how you don't need concurrency to have parallelism.
Re:Nuts and Bolts of Multithreaded Programming (2011)
by deepnotderp on Sunday July 02, 2017 @08:30PM
(#14685076)
Technically, most modern GPUs are SIMT meaning that they aggregate lockstep threads into vector instructions, so yes.
Re:Nuts and Bolts of Multithreaded Programming (2011)
by fizixer on Sunday July 02, 2017 @07:59PM
(#14684971)
I wouldn't think so. Not sure why you asked though.
Re:Nuts and Bolts of Multithreaded Programming (2011)
by nine_k on Sunday July 02, 2017 @10:34PM
(#14685521)
There's a question upthread asking why GPUs are not mentioned in "Nuts and bolts of multi-threaded programming". Hence my wonderment.
Re:Nuts and Bolts of Multithreaded Programming (2011)
by fizixer on Monday July 03, 2017 @12:08AM
(#14685766)
Then you didn't read the comment upthread fully.
It did acknowledge the title, and contrasted it with the tone of the article (the first half at least) which is way broader minus a mention of GPUs (the basics of parallel algorithms, parallel APIs).
Re:Nuts and Bolts of Multithreaded Programming (2011)
by mockery on Sunday July 02, 2017 @11:56PM
(#14685732)
But GPUs aren't SIMD. (In some sense they're almost the opposite.)
Re:Nuts and Bolts of Multithreaded Programming (2011)
by vilya on Monday July 03, 2017 @04:21AM
(#14686737)
In an incorrect sense, yes. :-)
SIMD = Single Instruction Multiple Data, meaning the same instruction being applied to multiple different values simultaneously. That's exactly what GPUs do.
Re:Nuts and Bolts of Multithreaded Programming (2011)
by telperion on Tuesday July 04, 2017 @11:00AM
(#14696218)
That's exactly what GPUs do in a single thread, but GPUs are also about threading, usually sporting tens, if not hundreds of small cores to effectively allow them to compute that many pixels (or rather shader outputs) in parallel. Otherwise, it wouldn't scale much.
That being said, I understand you only wanted to point out the error in the upper post.