Category: Intel Distribution for Python

Lean Code and the Intel Distribution for Python Performance

Python is one of my favourite languages for prototyping and developing applications.  Unfortunately, as an interpreted language, Python’s performance is of great concern when running an application at scale in a production environment.  Developers will typically resort to using languages like C or C++ that can be compiled into a high performance application for this very reason.

Fortunately, as I’ve stated many times before, “There’s always a better way and more than one solution!”  Although you can download the latest version of Python directly from the Python.org website, other distributions are available.  I use the Anaconda Python distribution as it is both comprehensive and easy to manage using Conda or Anaconda Navigator.

To overcome some of Python’s performance shortfalls you can “Supercharge applications and speed up core computational packages” with the INTEL Distribution for Python, powered by Anaconda.  The Intel Distribution for Python is intended to address performance constraints that would otherwise inhibit you from using your application in a production environment at scale.

This Anaconda & Intel Python Distribution FAQ presents the differences and performance gains.  Note that both distributions are available for Windows, MacOS, and Linux.

Install Anaconda3

Anaconda3 5.2.0 (64-bit) SetupDownload and install the latest version of the Anaconda Python Distribution (5.2.0 as of this writing).  The installation wizard will guide you through the installation process, allowing you to install for “Just Me” or “All Users” and will prompt you choose an install location.  In most cases, the default settings will suffice.

Anaconda3 5.2.0 (64-bit) Choose Install LocationWhen the installation is complete, you will be presented with an option to download Visual Studio Code.  Although I’m using PyCharm by JetBrains, Visual Studio Code is an excellent editor that I find myself using from time to time.  Anaconda’s Spyder is also an excellent IDE and is part of the Anaconda installation.

Note:  You will be asked to sign up for an Anaconda Cloud account.  This will enable you to manage and distribute your packages and upload and download your environments, notebooks, and projects Anaconda Cloud.

Install the Intel Distribution for Python

After you have successfully installed Anaconda on your machine, follow these step by step  instructions to take advantage of the Intel Distribution for Python in your production environment.  I highly recommend reading the instructions BEFORE proceeding with the step by step instructions provided as several installation options are presented and may vary depending on your operating system.  Note that you may elect to install either the core or full Intel Distribution for Python depending on your requirements and versions are available for both Python 2 and Python 3.

Anaconda Prompt From Windows Start MenuOn Windows 10, go to the start menu and scroll down to find the “Anaconda3” application.  From the drop down, click the “Anaconda Prompt”.  This will open a terminal window from which you can enter the applicable “conda” commands referenced in the “Intel Distribution for Python” installation instructions.

We opted to install the full Intel Distribution for Python, although the core Intel Distribution for Python may serve your immediate needs.  Since the Intel Distribution for Python is powered by Anaconda, you can easily manage your Python packages and environments accordingly.

Note that the Full Intel Distribution for Python is quite large and will take some time to download depending on the speed of your internet connection.

Having the ability to significantly increase the performance of your application by using Intel’s Distribution for Python is definitely far more productive, effective, and efficient than having to rewrite your application in another language for the sake of performance alone.

Until Next Time – STAY lean!
Versalytics

Related Articles and Resources

 

 

 

Advertisements

High-Performing Python

I was pleasantly surprised to see some very useful downloads when I opened the November 2017 issue of the Intel Developer Products Newsletter:

  • Intel Distribution for Python – Accelerate Python Performance – Powered by ANACONDA, and
  • Intel Performance Libraries – Download these free libraries today to create better, more reliable, and faster software applications.
    • Intel Data Analytics Acceleration Library (Intel DAAL)
    • Intel Math Kernel Library (Intel MKL)
    • Intel Integrated Performance Primitives (Intel IPP)
    • Intel Threading Building Block (Intel TBB)
    • Intel MPI Library (Windows Package)

Rather than write about these free offerings from Intel, click on the links above to see and choose the tools that will serve you best.  The files are relatively large and you may want to reserve your time for when and how you download them to your machine.  Versions are available for Windows, MacOS, and Linux.

Anaconda is my Python installation of choice if not simply for the ease of maintaining the many Python packages that are available.  That the Intel distribution for Python is powered by Anaconda is not surprising.

Although I’ve expressed my fair share of caution when using code libraries written by others, I have no concerns with Intel as the documentation is more than thorough.  If you’re truly interested in a comprehensive Python installation, consider this package offering from Intel.

Until Next Time, stay LEAN!

Versalytics