Talk:Parallel computing
<templatestyles src="Module:Message box/tmbox.css"/><templatestyles src="Talk header/styles.css" />
| This is the talk page for discussing improvements to the Parallel computing Template:Pagetype. This is not a forum for general discussion of the article's subject. |
Article policies
|
| Template:Find sources |
Script error: No such module "Check for unknown parameters".Script error: No such module "Check for deprecated parameters".
Template:Article history Template:WikiProject banner shell Template:Archives User:MiszaBot/config User:HBC Archive Indexerbot/OptIn
External links modified
Hello fellow Wikipedians,
I have just modified one external link on Parallel computing. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
- Added archive https://web.archive.org/web/20150128133120/https://www.nersc.gov/users/computational-systems/hopper/configuration/interconnect/ to https://www.nersc.gov/users/computational-systems/hopper/configuration/interconnect/
When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.
Cheers.—InternetArchiveBot (Report bug) 23:49, 7 November 2017 (UTC)
Template:WJS invitation 2018 T.Shafee(Evo&Evo)talk 01:13, 30 October 2018 (UTC)
GPU (SLI, Crossfire)
Similar principle was applied to GPUs: Scalable Link Interface, AMD CrossFireX, NVLink Bridge. It would be good to expand upon, how it is similar and how it differs from CPU parallel computing. ... It is based on the principle of parallel processing where two or more GPUs share the load... 89.201.184.8 (talk) 09:03, 26 April 2020 (UTC)
A more fundamental classification
I'm missing a more fundamental classification of parallelism that is not rooted in implementation details of contemporary computers, or computing at all, just the possible ways to split work.
I think there are fundamentally two ways to split work so it can be worked on simultaneously:
- Parallel: Distribute the work so that work units take different paths.
- Series: Split the work in stages that can be pipelined.
Someone must have thought about this before.—Anordal (talk) 21:40, 14 January 2021 (UTC)
Parallel vs. Concurrent
Based on the descriptions in these two articles, how can you be parallel without being concurrent? Digital27 (talk) 01:49, 15 November 2023 (UTC)
- On further thought, should we really be making the distinction between parallel and concurrent? Digital27 (talk) 16:54, 15 November 2023 (UTC)
- They are different.
- When two jobs run parallel they run at the same time.
- When two jobs run concurrent, they share a core but appear to run at the same time. This is multitasking. FrankYang43338 (talk) 18:44, 23 March 2024 (UTC)