oneAPI DevSummit for AI & HPC 2023, South Korea

November 2, 2023
2:00 - 5:00 KST
See agenda

DESCRIPTION

Join us for the oneAPI DevSummit, focused on AI, oneAPI and SYCL for accelerated computing across xPU architectures. In this virtual conference, you will learn from leading industry and academia speakers innovating new oneAPI solutions for cross-platform, multi-vendor architecture.

Why attend?
Join the global community of software innovators at the oneAPI DevSummit for AI & HPC. In this 1-day summit you will gain insights into:

  • Latest advancements shaping the future of high-performance computing and AI
    Opportunities to attend Keynotes, Customer voice, Lightning talks from industry-leading experts at Infosys Edgeverve, PrediQt, and Moasys.
  • Check out performance through Intel® AI analysis tool practice, and practice migrating and optimizing oneAPI CUDA to SYCL code

Agenda

Nov. 2nd

Nov. 2nd

2:00 - 2:05 KST Welcome & Introduction

Welcome & Introduction

Welcoming message with Intel’s investment of Software for developers including oneAPI and vision of Intel® AI technology, overall introduction of the event. Download Presentation
Presenting

2:05 - 2:10 KST Keynote

Keynote

Scott will be sharing the momentum of the oneAPI ecosystem across the world with some key examples of how the developer community has adopted oneAPI. He will also share the new initiatives from Intel® to help developers and start-ups learn and adopt oneAPI. He will also discuss the new capabilities within the oneAPI toolkits available in 2023. Download Presentation
Presenting

2:10 - 2:30 KST Customer Voice

Targeting NVIDIA Devices with Intel® oneAPI and SYCL

NVIDIA® GPUs and CUDA programming language has remained the standard in HPC and AI workloads forming a closed ecosystem. The arrival of diverse classes of accelerators have opened new opportunities for innovation across multiple industries. We introduce oneAPI, an open, cross-industry, standards-based, unified, multiarchitecture, multi-vendor programming model. oneAPI uses SYCL, an openly spec’d alternative to CUDA that supports many classes of accelerators like CPUs, GPUs,…
Presenting

2:30 - 2:50 KST Lightning Talk

MLM Fine-Tuning & Optimization with oneAPI and 4th Generation Intel® Xeon® Scalable Processor

In this talk, we present our experiences about using oneAPI with Intel® 4th Gen Xeon Sapphire Rapids processors to showcase how we saw almost 50% reduction in training time for fine-tuning an MLM model on domain data. We were able to train an MLM model with finance domain data containing ~1.1 M paragraphs, on Intel® 4th Gen Xeon Sapphire Rapids (with resource limitation of 32…
Presenting

2:50 - 3:10 KST Lightning Talk

Molecular Dynamics using SYCL and complete offload analysis

This project is a molecular dynamics simulation of N particles in a 3D box, with periodic boundary conditions. The particles interact with each other through the Lennard-Jones potential, which is a commonly used pairwise potential in molecular simulations. The simulation is implemented using the SYCL (Single-source DPC++ heterogeneous programming for OpenCL) programming model for parallelism. The simulation consists of two main steps: computing pairwise forces…
Presenting

3:10 - 3:25 KST Trivia Game - Quiz on oneAPI, SYCL, FPGA & AI

3:25 - 4:45 KST oneAPI Hands-on Tutorial - SYCL

oneAPI CUDA to SYCL code migration and optimization practice

In this session, we’ll show you two hands-on demonstrations implementing oneAPI. 1. Learn about Intel® oneAPI SYCLomatic, a tool that converts CUDA code to SYCL, and practice converting CUDA code (mini-nbody) to SYCL using SYCLomatic. Let’s look at performance by running on CPU (SPR) and GPU (PVC) in Intel® Developer Cloud. 2. Analyze the performance of the Fortran program that performs matrix calculations using the…
Presenting

3:25 - 4:45 KST Intel® AI Analytics Hands-on Tutorial - AI

Intel® AI Analytics Tools Practice and Performance Check Practice

This workshop session intends to help you understand Intel® AI Analytics Toolkit (Intel® AI Kit) for giving data scientists, AI developers, and researchers familiar Python* tools and frameworks to accelerate end-to-end data science and analytics pipelines on Intel® architecture. The contents of this presentation will show components built on oneAPI libraries for low-level compute optimizations. Moreover, by deep diving into optimization techniques that are applied…
Presenting

4:45 - 5:00 KST Closing

Closing

He will briefly summarize today’s events and answer some questions that would be beneficail to share. Lastly, he will announce the results of the contest to the participants.
Presenting
×


Register for the oneAPI DevSummit hosted by UXL:

Register Now