Python is both easy to learn and is backed up with vast range of open source packages and libraries. Now it gets a further performance upgrade thanks to Intel’s Parallel Studio XE 2019.

Richard Friedman tells more in this report from Inside HPC:

Industrial strength data analytics involves some very serious math. A single application might employ many complex solutions requiring a significant effort to develop. With the Intel Data Analytics Acceleration Library (Intel DAAL), the data scientist has all the building blocks they need, in Python, for all stages of data analysis from data acquisition through prediction and decision making. And, it scales from a single node to a large cluster with remote storage without additional effort.

At the center of all this, Intel DAAL is a highly optimized library of computationally intensive routines supporting best performance on Intel architectures including Intel Xeon processors, Intel Core processors, Intel Atom processors, and Intel Xeon Phi™ processors. Intel DAAL provides Python with a rich set of algorithms, ranging from the most basic descriptive statistics for datasets to more advanced data mining and machine learning algorithms.

Python programmers can easily utilize Intel DAAL (daal4py) for developing robust, scalable, high performing data processing right out of the box, and immediately take advantage of its features and. Shown to give a substantial performance boost over alternatives, data scientists programming in Python with Intel DAAL can implement batch, online, clustering, and much more within their Python applications.

In many cases, existing Python applications will perform significantly better merely by switching to the Intel distribution. On distributed parallel systems, Intel Python supports the mpi4py library, which interfaces the Intel MPI Library over InfiniBand and the Intel Omni-Path communications fabric. The result is decreased latency and increased scaling for distributed Python applications.