What is Intel recommended way to setup MKL, NumPy and Python

[swan_m_intel]

I am looking to set up MKL on a Linux Cascade lake Ubuntu 20.4 server. 
I would like to know Intel recommended way to install Python, NumPy and MKL.

Also, the recommended flags for best performance config. 

[Soumya_M_Intel]

Hi swan_m_intel,

Thanks for posting your concern on the forum. Please know the Intel® oneAPI Math Kernel Library & Intel® Distribution for Python are components of the Intel® oneAPI Base Toolkit.

You may refer to the below link for guidance over the installation of the toolkit and its standalone components.

Link: https://software.intel.com/content/www/us/en/develop/documentation/installation-guide-for-intel-oneapi-toolkits-linux/top/installation/install-using-package-managers/list-available-packages.html#list-available-toolkits-components-and-runtime-library-packages_list-available-packages

Please refer to the below link to find the supported compatibility for the product toolkit and its components. These are the tested and supported configurations for them.

Link: https://software.intel.com/content/www/us/en/develop/articles/intel-oneapi-base-toolkit-system-requirements.html

Have a nice day ahead.

Regards,

Soumya

[swan_m_intel]

If I do 'pip install mkl', will it pull all its dependencies?

Where can I find mkl dependency list if I have to go with the public version of python and Numpy?  Also do I need to do below mentioned line >> 

icc app.obj ${MKLROOT}/lib/intel64/libmkl_intel_lp64.so.1 ${MKLROOT}/lib/intel64/libmkl_intel_thread.so.1 ${MKLROOT}/lib/intel64/libmkl_core.so.1 -liomp5 -lpthread -lm -ldl

[Soumya_M_Intel]

Hi swan_m_intel,

Thanks for getting back to us on the post. We are currently checking on the mentioned queries with our internal team. Please wait for 2-3 business days. We will get back to you as soon as possible.

Regards,

Soumya

[Ruqiu_C_Intel]

Hi,

We haven't specific recommended flags for best performance config. But we do have one article about “Install NumPy/SciPy” which make them have good performance with MKL enabled.

Please check here:

https://software.intel.com/content/www/us/en/develop/articles/build-numpy-with-mkl-and-icc.html

[Ruqiu_C_Intel]

This issue is closing and we will no longer respond to this thread. If you require additional assistance from Intel, please start a new thread. Any further interaction in this thread will be considered community only.