42
edits
Changes
→Computation Modes
The final method of processing in the library is distributed processing. This is exactly what it sounds like, the library now forks different chunks of data to different compute nodes before finally rejoining all the data in one place.
[https://github.com/oneapi-src/oneDAL/blob/master/examples/daal/cpp/source/kmeans/kmeans_dense_distr.cpp K-means Code Example]
[https://github.com/oneapi-src/oneDAL/blob/master/examples/daal/cpp/source/moments/low_order_moms_dense_distr.cpp Moments of Low Order Example]
[https://github.com/oneapi-src/oneDAL/blob/master/examples/daal/cpp/source/pca/pca_cor_dense_distr.cpp Principle Component Analysis]
[https://github.com/oneapi-src/oneDAL/blob/master/examples/daal/cpp/source/svd/svd_dense_distr.cpp Distributed Processing]
[[File:alow7.jpg]]