Traduction et analyse de mots par intelligence artificielle ChatGPT
Sur cette page, vous pouvez obtenir une analyse détaillée d'un mot ou d'une phrase, réalisée à l'aide de la meilleure technologie d'intelligence artificielle à ce jour:
- comment le mot est utilisé
- fréquence d'utilisation
- il est utilisé plus souvent dans le discours oral ou écrit
- options de traduction de mots
- exemples d'utilisation (plusieurs phrases avec traduction)
- étymologie
Qu'est-ce (qui) est data-parallel - définition
![A graphical representation of [[Amdahl's law]]. The speedup of a program from parallelization is limited by how much of the program can be parallelized. For example, if 90% of the program can be parallelized, the theoretical maximum speedup using parallel computing would be 10 times no matter how many processors are used.](https://commons.wikimedia.org/wiki/Special:FilePath/AmdahlsLaw.svg?width=200 "A graphical representation of [[Amdahl's law]]. The speedup of a program from parallelization is limited by how much of the program can be parallelized. For example, if 90% of the program can be parallelized, the theoretical maximum speedup using parallel computing would be 10 times no matter how many processors are used.")
![Beowulf cluster]]](https://commons.wikimedia.org/wiki/Special:FilePath/Beowulf.jpg?width=200 "Beowulf cluster]]")
![Blue Gene/L]] massively parallel [[supercomputer]]](https://commons.wikimedia.org/wiki/Special:FilePath/BlueGeneL cabinet.jpg?width=200 "Blue Gene/L]] massively parallel [[supercomputer]]")
![The [[Cray-1]] is a vector processor](https://commons.wikimedia.org/wiki/Special:FilePath/Cray 1 IMG 9126.jpg?width=200 "The [[Cray-1]] is a vector processor")
![A graphical representation of [[Gustafson's law]]](https://commons.wikimedia.org/wiki/Special:FilePath/Gustafson.png?width=200 "A graphical representation of [[Gustafson's law]]")
![[[ILLIAC IV]], "the most infamous of supercomputers"<ref name="infamous"/>](https://commons.wikimedia.org/wiki/Special:FilePath/ILLIAC 4 parallel computer.jpg?width=200 "[[ILLIAC IV]], \"the most infamous of supercomputers\"<ref name=\"infamous\"/>")
![A logical view of a [[non-uniform memory access]] (NUMA) architecture. Processors in one directory can access that directory's memory with less latency than they can access memory in the other directory's memory.](https://commons.wikimedia.org/wiki/Special:FilePath/Numa.svg?width=200 "A logical view of a [[non-uniform memory access]] (NUMA) architecture. Processors in one directory can access that directory's memory with less latency than they can access memory in the other directory's memory.")
![Tesla GPGPU card]]](https://commons.wikimedia.org/wiki/Special:FilePath/NvidiaTesla.jpg?width=200 "Tesla GPGPU card]]")
![Taiwania 3 of [[Taiwan]], a parallel supercomputing device that joined [[COVID-19]] research.](https://commons.wikimedia.org/wiki/Special:FilePath/Taiwania 3 Supercomputer.jpg?width=200 "Taiwania 3 of [[Taiwan]], a parallel supercomputing device that joined [[COVID-19]] research.")
Wikipédia
Data parallelism is parallelization across multiple processors in parallel computing environments. It focuses on distributing the data across different nodes, which operate on the data in parallel. It can be applied on regular data structures like arrays and matrices by working on each element in parallel. It contrasts to task parallelism as another form of parallelism.
A data parallel job on an array of n elements can be divided equally among all the processors. Let us assume we want to sum all the elements of the given array and the time for a single addition operation is Ta time units. In the case of sequential execution, the time taken by the process will be n×Ta time units as it sums up all the elements of an array. On the other hand, if we execute this job as a data parallel job on 4 processors the time taken would reduce to (n/4)×Ta + merging overhead time units. Parallel execution results in a speedup of 4 over sequential execution. One important thing to note is that the locality of data references plays an important part in evaluating the performance of a data parallel programming model. Locality of data depends on the memory accesses performed by the program as well as the size of the cache.
