4 ms·
You're talking about data parallelism, the OP is talking about task parallelism. Since in both cases things are occurring in parallel it makes sense to use 'pa
by Symmetry 10y ago
You're talking about data parallelism, the OP is talking about task parallelism. Since in both cases things are occurring in parallel it makes sense to use 'parallelism' as an umbrella term for both and it was clear from context here that it was task parallelism under discussion.
- rbehrends 10y agoI am talking about both. You can have task parallelism in a distributed system with no shared memory, after all.