oneAPI: A Unified Programming Model based on C++ and SYCL

oneAPI: Cross Architecture Programming
oneAPI: Cross Architecture Programming
Public group

Intel San Jose Innovation (SJI) Campus

131 Innovation Drive · San Jose, ca

How to find us

We will be in Bldg 3 (where the pin is in map) in the SJI-3 Auditorium. Parking is available in front of bldg 3 or 4, or in the parking garage (just below building 4 in the map).

Location image of event venue

Details

Register Today: http://www.cvent.com/d/bhqsr2

If you are a programmer who wants to build and run a compute-intensive, data-centric application or workload, such as HPC, AI, data analytics and more, across multiple architectures including CPUs, GPUs, FPGAs and other accelerators, you know what a nightmare it is to learn all the different programming models and make them work together even before you start optimizing anything. oneAPI, whose backbone is Data Parallel C++ (DPC++) based on C++ and SYCL, makes cross-architecture development an easier task. Join us to learn about oneAPI, a unified software programming model aimed to simplify heterogeneous computing:

1. Learn how oneAPI streamlines coding across diverse architectures, with highlights of specific examples
2. Dive into DPC++ best practices, based on C++ and SYCL
3. Discover how to find and optimize performance bottlenecks across multiple architectures
4. Learn how to develop new deep learning frameworks—or customize existing ones—on CPUs and GPUs for single-node or multi-node distributed processing using the Intel® oneAPI DL Framework Developer Toolkit
5. Find out how to more easily program FPGAs using the Intel® FPGA Add-on for the Intel® oneAPI Base Toolkit
6. Get started quickly and test your code in the Intel® DevCloud with a walkthrough of example DPC++ code, along with a few of the Intel® oneAPI Beta Toolkits

View Speaker Details and Abstracts:
http://www.cvent.com/d/bhqsr2/6X

Agenda:
1:00pm Welcome
1:05pm oneAPI Overview and Workshop
2:35pm Break
2:50pm Using Intel® Advisor and Intel® VTune™ Profiler to Guide Performance Optimization
3:35pm Performance Optimization via Intel® oneAPI DL Framework Developer Toolkit
4:35pm Development Workflow and Tools for Targeting FPGAs with oneAPI
5:05pm Dinner Break
5:35pm Evening Welcome
5:40pm Immersive Experiences: Poised to Revolutionize How We Interact with Our World
6:10pm Migration of GPGPU HPC Applications to oneAPI Data Parallel C++
7:00pm Lightning Talks
7:40pm Raffle

---------
2-Step Registration Process:
1. Click the "Are you going?" button here on meetup. This means you'll get any relevant info on this meetup including agenda, changes to time or location, and also get updated when the presentations are available after the event.

2. Register for the event at http://www.cvent.com/d/bhqsr2
This will make check-in a breeze. If you miss this step, you'll need to register at the door when you arrive.