oneAPI DevSummit for AI and HPC 2023

December 4, 2023
12/4/23: 9:00 - 10:30 am CST 12/5/23: 9:00 - 4:00 pm CST 12/6/23: 9:00 - 3:15 pm CST
See agenda

DESCRIPTION

Reserve your FREE spot at the 2023 oneAPI DevSummit for AI and HPC 2023 to explore cross-architecture programming, AI, HPC, and more. Learn from industry experts, discover oneAPI advancements, gain insights into performance analysis tools, and experience hands-on workshops with Intel Developer Cloud. Join speakers from prestigious organizations and get inspired by keynotes, tech talks, and tutorials. Elevate your developer journey with a focus on AI and HPC, including data preparation, training, inference, deployment, and scaling.

Don’t miss this opportunity to join renowned experts for deep dives into cross-architecture software development spanning:

  • AI and analytics using industry-standard frameworks and tools, performance-optimized by oneAPI
  • What’s new with oneAPI, including the Intel® AI Analytics and HPC Toolkits
  • Overviews of key performance analysis tools, including how to use them and where to get them
  • The latest on the oneAPI specification
  • Hands-on workshops using Intel(r) Developer Cloud

Agenda

December 4th
December 5th - AI
December 6th - HPC

December 4th

9:00 - 10:30 CST

Intel® Developer Cloud and Discord Onboarding Day

Prepare for the DevSummit by attending day zero, Onboarding Day! During this day we will help onboard you onto Discord as well as the Intel(r) Developer Cloud. Attendees will need to sign up for the Intel(r) Developer Cloud in order to participate in the hands-on training sessions. Join us to ask questions, play games, and have fun during this registration party!
Presenting

December 5th - AI

9:00 - 9:15 CST

Introduction

Presenting

9:15 - 10:00 CST Keynote

Democratizing the use of AI: FUJITSU - MONAKA

Artificial intelligence (AI) is rapidly evolving, but accessibility remains a challenge. Our mission is to democratize AI, fostering open collaboration with the global community to develop human-centered, sustainable digital transformation. Japan’s NEDO has launched an initiative to achieve 40% energy savings in domestic data centers by 2030. Fujitsu’s MONAKA, a 2 nm Arm CPU arriving in 2027, focuses on energy efficiency to support carbon-neutral green…
Presenting

10:05 - 10:35 CST Tech Talk

Leveraging oneAPI Containers to Deploy OpenVINO™ Notebooks for Efficient AI Research

This presentation will demonstrate how to use Apptainer containers to deploy Intel OpenVINO Jupyter notebooks for an optimized AI research workflow. OpenVINO is Intel’s toolkit for deploying pre-trained deep learning models and optimizing model inference. The oneAPI containers package all the required software dependencies in a self-contained unit. Apptainer enables seamless integration of these containers into HPC clusters such as Digital Research Alliance of Canada,…
Presenting

10:35 - 10:50 CST Break

10:50 - 11:20 CST Tech Talk

Enhancing Aviation Predictive Maintenance with oneAPI: NASA's Turbofan Engine Degradation Simulation

NASA’s Open Data Portal facilitates Earth science research, climate monitoring, disaster management, and international collaboration. The CMAPSS simulated data, depicting aircraft engine degradation, is pivotal for predictive maintenance, research, and education in aviation and data science. This dataset describes the degradation of the aircraft and its contamination with sensor noise. Learn how Deepthi harnessed oneAPI’s unified programming model, integrating sklearn, xgboost, and daal4py libraries, to…
Presenting

11:25 - 12:30 CST Workshop

Gen AI: Unleash the Power of GenAI and LLM Using Latest Intel Cloud GPU

In this talk, I will demonstrate how GenAI can be used to develop a complete backend for a prompt generation and API integration application. This application will be able to generate prompts for a variety of tasks, such as writing emails, creating social posts, and developing marketing materials. Text-to-image stable diffusion models are a new generation of generative AI models that can generate photorealistic images…
Presenting

12:30 - 1:00 CST Lunch

1:00 - 1:20 CST Lightning Talk

Learn more about the Unified Acceleration (UXL) Foundation

The newly launched UXL Foundation, part of the Linux Foundation JDF Family, is building a new multi-vendor and cross architecture accelerator software ecosystem. Come join us to learn more about governance, the specification, SIGs and open source projects and how you can get involved in driving the future of open accelerated computing. Download Presentation
Presenting

1:25 - 1:55 CST Tech Talk

Porting SYCL Accelerated Neural Network Frameworks to Edge Devices

Portable hardware acceleration has become increasingly necessary with the rise of the popularity of edge computing, that encourages data to be processed and stored as close to the source of origination as possible. This capability is needed in areas where bandwidth and latency are restricted and network stability, privacy, or security are unreliable or insecure. A representative use case is unmanned aerial vehicles (UAVs) where…
Presenting

1:55 - 2:05 CST Break

2:05 - 2:35 CST Tech Talk

Redefining Voice Signal Processing with Habana Gaudi

Voice signal processing stands pivotal in AI’s progression, necessitating dynamic computational solutions. Enter Habana Gaudi: a paradigm in AI processors promising unparalleled efficiency and scalability. With voice signals’ intricate time-frequency dynamics, Gaudi’s integration with HBM2 offers optimized memory access while its support for TensorFlow and PyTorch bridges hardware power with software finesse. Emphasizing scalability, rapid memory access, and performance efficiency, Gaudi’s architecture is primed for…
Presenting

2:40 - 3:40 CST Demo

Intel® Gaudi® in Action: Solving Real-World Challenges with Fine-Tuned Language Models

Join our session for practical insights and technical expertise perfect for developers eager to leverage advanced language capabilities for specific industries. Burak from Spiky will kick off by discussing how real-time sales playbooks benefit from language models. Learn how Spiky utilizes Intel Gaudi 1 DL1 instances to fine-tune Large Language Models (LLMs) on datasets like sales call transcripts and industry jargon, improving sales correspondence language.…
Presenting

3:40 - 4:00 CST

Conclusion

Presenting

December 6th - HPC

9:00 - 9:15 CST

Introduction

Presenting

9:15 - 10:00 CST Keynote

Reflect, Rejoice, Envision: Navigating this year’s oneAPI Journey

During “Reflect, Rejoice, Envision: Navigating the Journey of a Year of oneAPI Together,” we will embark on an inspiring exploration of the past year’s accomplishments, challenges, and milestones. This keynote invites you to join us as we delve into the pivotal moments that shaped 2023 for the oneAPI community. As we reflect on the year, we celebrate the community, programs, partners, and contributors to the…
Presenting

10:05 - 10:35 CST Tech Talk

Development and Optimization of a SYCL Backend for libCEED

libCEED is a new library, developed as part of the Exascale Computing Project (ECP), that provides a highly efficient and portable interface for high-order finite element implementations that can be easily integrated with a wide variety of scientific software based on discretization. The libCEED library provides high-performance, GPU specific implementations using CUDA, HIP, and MAGMA backends that can be chosen at runtime by the source…
Presenting

10:40 - 11:10 CST Tech Talk

Performant Portable HPC Applications with SYCL and oneMKL Interface

With the recent emergence of numerous accelerators promising large-scale computational speed-ups over traditional CPU-based applications, heterogeneous computing has become of increasing interest to the scientific computing community. However, producing portable code without sacrificing the performance available from vendor-specific libraries remains a challenge, for example, when using discrete Fourier transforms (DFTs). An example of this is GROMACS, which must consequentially maintain multiple GPU programming APIs and…
Presenting

11:10 - 11:20 CST Break

11:20 - 11:50 CST Tech Talk

Energy-Efficient Heterogenous Computing with SYNERGY

Energy-efficient computing aims to maximize the energy efficiency of computing systems. A key approach in modern parallel hardware is frequency scaling, enabled by Dynamic Voltage and Frequency Scaling. AMD, Intel, and NVIDIA GPUs provide APIs for power management and core frequency adjustment. Leveraging these capabilities allows optimizing both performance and energy consumption in GPU programs. In this presentation, Biagio will introduce SYnergy, a novel energy-efficient…
Presenting

11:55 - 12:25 CST Tech Talk

Building a Portable, Scalable, Performant ZFP Backend Using oneAPI and SYCL to Advance Exascale Computing: A Developer Perspective

ZFP is a leading software for lossless and error-controlled lossy compression of floating-point data that is vital in high-performance computing. It is one of the software technologies explicitly selected by the DOE Exascale Computing Project (ECP) since its inception in 2016 because of its numerous science and engineering applications and users. This project aims to develop a portable, scalable, and performant ZFP backend to Intel’s…
Presenting

12:30 - 1:00 CST Tech Talk

Early Results Using Fortran's `Do Concurrent` Standard Parallelism on Intel GPUs with the IFX Compiler

Writing GPU-accelerated code often requires specialized language extensions or directive APIs like OpenACC or OpenMP. Using features within standard languages to achieve GPU-acceleration has gained interest, as it allows for portable code that is easier to develop and understand, and has increased (vendor independent) longevity. For our production Fortran codes, we have successfully used the language’s `Do Concurrent` (DC) construct for GPU acceleration on NVIDIA…
Presenting

1:00 - 1:30 CST Lunch

1:30 - 2:30 CST Panel

The Journey to Ubiquitous SYCL Adoption

In this panel discussion, we will talk with experts and software development leaders in research and industry about their path towards adopting the open cross-platform SYCL abstraction layers. We will explore the benefits of SYCL for an open multi architecture, open accelerated compute ecosystem. Not only will we touch on what makes adoption of and migration to SYCL easy, we will also try to understand…
Presenting

2:35 - 3:05 CST Tech Talk

SYCL Support for Continental-Scale Ecological Observations: Scalable and Portable Blending of Massive Image Mosaics

The National Ecology Observatory Network (NEON) is a continental-scale facility which gathers hundreds of terabytes of long-term open-access ecological data to study changes in US ecosystems. Most of this data comes from Airborne Observation Platforms (AOPs) and it is processed and shared with the scientific community through OpenViSUS, developed by the University of Utah and deployed by VISOAR LLC. Some data analysis components of this…
Presenting

3:05 - 3:15 CST

Conclusion

Presenting
×


Learn about joining the UXL Foundation:

Join now