CUDA DLI TRAINING SESSIONS AT GTC 2019
EXPECTED TO BE THE BIGGEST YET, GTC
FEATURES SESSIONS AND DLI TRAINING ON THE
MOST IMPORTANT TOPICS IN COMPUTING TODAY
WHY DLI HANDS-ON TRAINING?
● LEARN HOW TO BUILD APPS ACROSS INDUSTRY SEGMENTS
● GET HANDS-ON EXPERIENCE USING INDUSTRY-STANDARD SOFTWARE, TOOLS & FRAMEWORKS
● GAIN EXPERTISE THROUGH CONTENT DESIGNED WITH INDUSTRY LEADERS
FUNDAMENTALS OF ACCELERATED COMPUTING WITH CUDA
PYTHON
This course explores how to use Numba—the just-in-
time, type-specializing Python function compiler—
to accelerate Python programs to run on massively
parallel NVIDIA GPUs. You’ll learn how to:
● Use Numba to compile CUDA kernels from
NumPy universal functions (ufuncs)
● Use Numba to create and launch custom CUDA
kernels
● Apply key GPU memory management
techniques
ADD TO MY SCHEDULE
ADD TO MY SCHEDULE
The CUDA computing platform enables acceleration of
CPU-only applications to run on the world's fastest
massively parallel GPUs. Learn how to accelerate C/C++
applications by:
● Exposing the parallelization of CPU-only
applications, and refactoring them to run in parallel
on GPUs
● Successfully managing memory
● Utilizing CUDA parallel thread hierarchy to further
increase performance
ACCELERATING APPLICATIONS WITH CUDA C/C++
CUDA ON DRIVE AGX
Explore how to write CUDA code and run it on
DRIVE AGX. You'll learn about:
ADD TO MY SCHEDULE
● Hardware architecture of DRIVE AGX
● Memory Management of iGPU and dGPU
● GPU acceleration for inferencing
ACCELERATING DATA SCIENCE WORKFLOWS WITH
RAPIDS
The open source RAPIDS project allows data scientists
to GPU-accelerate their data science and data
analytics applications from beginning to end, creating
possibilities for drastic performance gains and
techniques not available through traditional CPU-only
workflows. Learn how to GPU-accelerate your data
science applications by:
ADD TO MY SCHEDULE
● Utilizing key RAPIDS libraries like cuDF & cuML
● Learning techniques and approaches to end-to-
end data science
● Understanding key differences between CPU-
driven and GPU-driven data science
DEBUGGING AND OPTIMIZING CUDA APPLICATIONS
WITH NSIGHT PRODUCTS ON LINUX TRAINING
Learn how NVIDIA tools can improve development
productivity by narrowing down bugs and spotting
areas of optimization in CUDA applications on a Linux
x86_64 system.
Through a set of exercises, you'll gain hands-on
experience using NVIDIA's new Nsight Systems and
Nsight Compute tools for debugging, narrowing down
memory issues, and optimizing a CUDA application.
ADD TO MY SCHEDULE
ACCELERATED DATA SCIENCE PIPELINE WITH
RAPIDS ON AZURE
Learn how to deploy RAPIDS machine learning jobs
on NVIDIA's GPUs using Microsoft Azure and
explore:
ADD TO MY SCHEDULE
● Azure Portal Permits: a convenient way to
perform functional experimentation with RAPIDS.
● Azure Machine Learning (AML) SDK: enables a
batch experimentation mode and where the user
can set ranges on different parameters to be run
on a RAPIDS program, saving the results for later
analysis
HIGH PERFORMANCE COMPUTING USING CONTAINERS
Learn to build, deploy and run containers in an HPC
environment.
During this session, you will learn: the basics of
building container images with Docker and Singularity,
how to use HPC Container Maker (HPCCM) to make it
easier to build container images for HPC applications,
and how to use containers from the NGC with
Singularity.
ADD TO MY SCHEDULE
INTRODUCTION TO CUDA PYTHON WITH NUMBA
Explore an introduction to Numba, a just-in-time
function compiler that allows developers to utilize
the CUDA platform in Python applications. You'll
learn how to:
ADD TO MY SCHEDULE
● Decorate Python functions to be compiled by
Numba
● Use Numba to GPU accelerate NumPy ufuncs
CUDA PROGRAMMING IN PYTHON WITH NUMBA AND
CUPY
Combining Numba, an open source compiler that can
translate Python functions for execution on the GPU,
with the CuPy GPU array library, a nearly complete
implementation of the NumPy API for CUDA, creates a
high productivity GPU development environment.
Learn the basics of using Numba with CuPy, techniques
for automatically parallelizing custom Python functions
on arrays, and how to create and launch CUDA kernels
entirely from Python.
ADD TO MY SCHEDULE
REGISTER TODAY FOR GTC
AND EXPLORE THE FULL
LIST OF CUDA TRAINING,
TALKS & EXPERT SESSIONS
LEARN MORE

CUDA DLI Training Courses at GTC 2019

  • 1.
    CUDA DLI TRAININGSESSIONS AT GTC 2019
  • 2.
    EXPECTED TO BETHE BIGGEST YET, GTC FEATURES SESSIONS AND DLI TRAINING ON THE MOST IMPORTANT TOPICS IN COMPUTING TODAY
  • 3.
    WHY DLI HANDS-ONTRAINING? ● LEARN HOW TO BUILD APPS ACROSS INDUSTRY SEGMENTS ● GET HANDS-ON EXPERIENCE USING INDUSTRY-STANDARD SOFTWARE, TOOLS & FRAMEWORKS ● GAIN EXPERTISE THROUGH CONTENT DESIGNED WITH INDUSTRY LEADERS
  • 4.
    FUNDAMENTALS OF ACCELERATEDCOMPUTING WITH CUDA PYTHON This course explores how to use Numba—the just-in- time, type-specializing Python function compiler— to accelerate Python programs to run on massively parallel NVIDIA GPUs. You’ll learn how to: ● Use Numba to compile CUDA kernels from NumPy universal functions (ufuncs) ● Use Numba to create and launch custom CUDA kernels ● Apply key GPU memory management techniques ADD TO MY SCHEDULE
  • 5.
    ADD TO MYSCHEDULE The CUDA computing platform enables acceleration of CPU-only applications to run on the world's fastest massively parallel GPUs. Learn how to accelerate C/C++ applications by: ● Exposing the parallelization of CPU-only applications, and refactoring them to run in parallel on GPUs ● Successfully managing memory ● Utilizing CUDA parallel thread hierarchy to further increase performance ACCELERATING APPLICATIONS WITH CUDA C/C++
  • 6.
    CUDA ON DRIVEAGX Explore how to write CUDA code and run it on DRIVE AGX. You'll learn about: ADD TO MY SCHEDULE ● Hardware architecture of DRIVE AGX ● Memory Management of iGPU and dGPU ● GPU acceleration for inferencing
  • 7.
    ACCELERATING DATA SCIENCEWORKFLOWS WITH RAPIDS The open source RAPIDS project allows data scientists to GPU-accelerate their data science and data analytics applications from beginning to end, creating possibilities for drastic performance gains and techniques not available through traditional CPU-only workflows. Learn how to GPU-accelerate your data science applications by: ADD TO MY SCHEDULE ● Utilizing key RAPIDS libraries like cuDF & cuML ● Learning techniques and approaches to end-to- end data science ● Understanding key differences between CPU- driven and GPU-driven data science
  • 8.
    DEBUGGING AND OPTIMIZINGCUDA APPLICATIONS WITH NSIGHT PRODUCTS ON LINUX TRAINING Learn how NVIDIA tools can improve development productivity by narrowing down bugs and spotting areas of optimization in CUDA applications on a Linux x86_64 system. Through a set of exercises, you'll gain hands-on experience using NVIDIA's new Nsight Systems and Nsight Compute tools for debugging, narrowing down memory issues, and optimizing a CUDA application. ADD TO MY SCHEDULE
  • 9.
    ACCELERATED DATA SCIENCEPIPELINE WITH RAPIDS ON AZURE Learn how to deploy RAPIDS machine learning jobs on NVIDIA's GPUs using Microsoft Azure and explore: ADD TO MY SCHEDULE ● Azure Portal Permits: a convenient way to perform functional experimentation with RAPIDS. ● Azure Machine Learning (AML) SDK: enables a batch experimentation mode and where the user can set ranges on different parameters to be run on a RAPIDS program, saving the results for later analysis
  • 10.
    HIGH PERFORMANCE COMPUTINGUSING CONTAINERS Learn to build, deploy and run containers in an HPC environment. During this session, you will learn: the basics of building container images with Docker and Singularity, how to use HPC Container Maker (HPCCM) to make it easier to build container images for HPC applications, and how to use containers from the NGC with Singularity. ADD TO MY SCHEDULE
  • 11.
    INTRODUCTION TO CUDAPYTHON WITH NUMBA Explore an introduction to Numba, a just-in-time function compiler that allows developers to utilize the CUDA platform in Python applications. You'll learn how to: ADD TO MY SCHEDULE ● Decorate Python functions to be compiled by Numba ● Use Numba to GPU accelerate NumPy ufuncs
  • 12.
    CUDA PROGRAMMING INPYTHON WITH NUMBA AND CUPY Combining Numba, an open source compiler that can translate Python functions for execution on the GPU, with the CuPy GPU array library, a nearly complete implementation of the NumPy API for CUDA, creates a high productivity GPU development environment. Learn the basics of using Numba with CuPy, techniques for automatically parallelizing custom Python functions on arrays, and how to create and launch CUDA kernels entirely from Python. ADD TO MY SCHEDULE
  • 13.
    REGISTER TODAY FORGTC AND EXPLORE THE FULL LIST OF CUDA TRAINING, TALKS & EXPERT SESSIONS LEARN MORE