O Edmond and Lily Safra International Institute of Neuroscience (IIN-ELS) and the NVIDIA[Company Name], a world leader in computer graphics and artificial intelligence, have partnered to hold a special event. A hands-on training session on the topics of... high performance computing and artificial intelligence It will be developed on the day July 25, 2018, aimed at students of Postgraduate Program in Neuroengineering.
The event will be held in Macaíba (RN), at the IIN-ELS headquarters, and will include two distinct parts:
Hands-On Lab: “OpenACC – 2X in 4 steps”
Hours: 9:30 AM – 12:00 PM
Participants will learn how to accelerate C/C++ or Fortran software using OpenACC to take full advantage of the massively parallel architecture of NVIDIA GPUs. OpenACC is a compilation directive-based approach, where the user provides hints to the compiler to accelerate their code, rather than writing the accelerator code themselves.
In 90 minutes, a four-step process for accelerating applications using OpenACC will be shown:
– Characterization and definition of application profiles;
Where and how to add compilation directives;
– Add directives to optimize data movement;
– Further optimize your code by using kernel parameterization.
Presentation: "New NVIDIA platform for HPC and Artificial Intelligence"
Time: 2 PM – 3:30 PM
Deep Learning (DL) is the Machine Learning (ML) technique that has been providing advancements in various industrial, commercial, and scientific workflows. NVIDIA's new Artificial Intelligence platform, comprised of hardware and software, is providing the computational power required by recent advancements in Deep Learning.
Volta, the company's newest and most advanced GPU architecture, was specifically designed for the high-performance computing workloads required for training and inferring Deep Neural Networks with massive amounts of training data. It is the first GPU architecture to include Tensor Cores (TCs), processing units designed for ultra-high-speed Tensor operations.
The latest version of the CUDA language (version 9) and NVIDIA SDKs have been enhanced to include specialized and highly optimized algorithms to extract the full potential of GPUs in DNN training and inference tasks across all Deep Learning Frameworks such as TensorFlow, CNTK, Caffe, etc. A wide variety of data can be efficiently used for training, including text, audio, images, and video. This new computing model is delivering excellent results in Computer Vision, Natural Language Processing, Language Translation, Speech Recognition, Recommendation Systems, Logistics, Autonomous Cars, and Robotics.