This page provides links to the current ways to get the Intel® Performance Libraries: Intel® oneAPI Math Kernel Library (oneMKL), Intel® Integrated Performance Primitives (Intel® IPP), or Intel® Data Analytics Acceleration Library (Intel® DAAL).
• Sub-component of Intel® oneAPI Base & HPC Toolkit or Intel® System Studio
After installing the suite package, the Intel Performance Libraries will be installed by default. The article Which version of the Intel IPP, oneMKL and Intel® Threading Building Blocks (Intel® TBB) Libraries are Included in the Intel® Composer Bundles? provides version information for the Intel Performance Libraries included in the bundled products:
• Free standalone access through separate product main pages:
• Free access through Intel Performance Libraries page (oneMKL, Intel IPP, Intel DAAL)
• YUM/APT repository for oneMKL, Intel IPP, Intel DAAL, and Intel® Distribution for Python* (for Linux*):
- Installing Intel® Performance Libraries and Intel® Distribution for Python* Using YUM Repository
- Installing Intel® Performance Libraries and Intel® Distribution for Python* Using APT Repository
• Conda* package/ Anaconda Cloud* support (oneMKL, Intel IPP, Intel DAAL, Intel Distribution for Python):
• NuGet distribution channel (oneMKL, Intel IPP, Intel DAAL, Intel TBB, oneMPI Library, Intel OpenMP):
- Installing Intel® Performance Libraries Using the NuGet Channel
- For Known Limitations, please visit the above link.
• PIP Install from the Python Package Index (PyPI) using pip (oneMKL, Intel IPP, Intel DAAL)
Only for oneMKL:
Cloudera* Parcels support since oneMKL 2017 update 2
Only for Intel IPP cryptography package:
- The Intel IPP cryptography package is an export-controlled product, and access requires a special request process. Please see Where do I download the Intel IPP Cryptography libraries? to download the Intel IPP cryptography libraries.
- The cryptography library in Intel IPP is also available as open source.
Only for Intel DAAL Java* package:
- Maven* Java distribution is available in Maven Central. Please see Intel DAAL Maven Release.
Only for pyDAAL module:
The libraries obtained through any of the above means are exactly the same in functionality and performance. The main differences are in license agreement and support priority. Please refer to the License FAQ.