A unique technical workshop for software developers, engineers and advanced data scientists to learn how to use the Intel® Software Development Tools to create, tune and optimize C/C++, Python and Fortran code for maximum performance on the latest Intel® Processors and Technologies.
Prerequisite: Please bring your own laptop / notebook. Some of the sessions will provide you the opportunity to experience the tools hands-on using your browser connected to Cloud instances pre-configured for your personal use.
After a quick introduction to the latest Intel® processors and technologies, this session give an overview of the Intel® Software Development Tools for High Performance Computing, Artificial Intelligence, Machine Learning / Deep Learning and Edge / IoT programming, including Intel® Parallel Studio XE and Intel® System Studio. You will get an in-depth overview of these comprehensive suites of development tools used to build and modernize code with the latest optimized libraries for C/C++, Fortran and Python, as well as best practice for code vectorization, multithreading, multinode parallelization, performance tuning and optimization on Intel® platforms.
In this lab we show, step by step, how you can modify a sample application through different optimization stages spanning through scalar tuning, vectorization tuning, memory tuning and threading tuning. The sample code used is an N-body simulation, which models how particles behave over time under the influence of the forces of gravity. The techniques used here can be readily applied to any applications you are developing.
Intel Advisor is a powerful tool for tracking down and solving vectorization problems. This demo will introduce Intel Advisor and especially how to use the survey and the trip count analyses to track all the performance issues of the previously introduced N-body simulation code. We will explain how to read and interpret Advisor's outputs to improve the vectorization and introduce the roofline analysis
This session, after first introducing the Intel® AI Software Stack for Developers & Data Scientists as well as the Intel® Distribution for Python (including NumPy / SciPy), will cover and demonstrate Intel’s technical best practices for developing and deploying Machine Learning and Deep Learning training on Intel CPU based platforms.
In this session we will introduce the Intel® Distribution of OpenVINO™ Toolkit for Computer Vision...
Based on convolutional neural networks (CNN), the toolkit extends workloads across Intel® hardware (including accelerators) and maximizes performance with deployment on Intel CPU, GPU, Movidius, and Intel FPGAs.
