-
Track #1
leveraging HPC and AI for Upstream
-
Track #2
Industrial Metaverse and AI for Oil, Gas, & Power Facilities
-
Track #3
ML Ops, IT Ops, and Architecture (HPC/AI)
-
Track #4
AI in Healthcare and Life Sciences
Energy Studio: Impact is a cutting-edge, web-based platform that harnesses big data to generate real-time energy analytics across the entire energy value chain. As the market becomes increasingly sophisticated, Energy Market Participants seek technology and analytics solutions that drive efficiencies and improve outcomes. To gain a complete view of the market, a convergence of datasets is required. Companies must be able to make smarter data-driven decisions in real-time to mitigate risk and gain a competitive edge. Energy Studio: Impact allows users to achieve key workflows in seconds and minutes rather than days or weeks, saving them valuable time.
How can we leverage advances in physics-inspired machine learning for a step change in field development design? How can they allow us to de-risk, accelerate planning cycles and make better decisions?
In this talk, we discuss new applications of AI and how they can integrate with existing tools to give high-performance and high-confidence solutions. A case study of field development for a CCS reservoir is discussed.
ExxonMobil has a decades long legacy of utilizing high performance computing (HPC) to solve large scale geoscience and engineering problems. Our supercomputing infrastructure serves multiple purposes including daily production use and as a platform for R&D. While our compute demand is historically dominated by seismic processing/imaging and reservoir engineering, we enable a broad range compute intensive applications to support our entire business. To maintain our HPC capability and make informed investment decisions, ExxonMobil is an active participant in the HPC community engaging with vendors, academia, and monitoring the competitor landscape. In this talk we will discuss our latest HPC investment in 2022, Discovery 5, which marks a significant change in our architectural direction from CPU-based systems to a GPU accelerated system based on NVIDIA GPUs. We will highlight business drivers informing this investment, answer the questions "Why GPUs" and "Why now", and review our software engineering and change management planning and execution strategy.
Automating salt interpretation in Gulf of Mexico (GoM) is a significant challenge. Various factors such as shape, roughness, size, composition, depth, and the type of seismic data acquisition, vintage, and imaging algorithm can impact image resolution, texture, and artifacts. However, using seismic images with similar processing, vintage, and acquisition geometry makes it possible to create machine learning (ML) models explicitly tailored for regional salt interpretation.
Using salt interpretation as a motivating application, we show how several innovations in data engineering and deep learning that enables our workflow. We will discuss how MDIO, an open-source format for storing multidimensional energy data enables us to build efficient training and inference pipelines to run experiments both on on-prem datacenters and cloud.
In this talk, we show an example on building a generalized salt interpretation model in the eastern Gulf of Mexico.
What if trash could transform into jet fuel, or a plastic bottle could be recycled forever? In this talk, we’ll explore how the development of sustainable processes has been augmented by recent advancements in three key areas: GPU technology, computing accessibility, and multiphase computational fluid dynamics software.
This presentation will highlight how CPFD Software’s Barracuda Virtual Reactor leverages the full NVIDIA stack to accelerate the development and commercialization of technology across a variety of industries. Case studies demonstrate outcomes including reduced time-to-market for new technologies, optimization of existing processes, and how previously intractable problems are now possible to solve.
Building High Fidelity (Cyber-Physical) Digital Twins (HFDT) is very complex and require highly advanced tools. NVIDIA Modulus and Omniverse present the perfect ecosystem. Softserve and NVIDIA are working closely on developing HFDT solutions. In this presentation we will discuss the state-of-the-art in Digital Twins, applications in the Oil and Gas industry, Modulus and Omniverse capabilities, ongoing Softserve/NVIDIA efforts using Modulus and Omniverse, and Softserve capabilities.
Bentley Systems is an infrastructure engineering software company that provides innovative software, including digital twins, to help organizations design, construct and operate their energy infrastructure. There are many benefits to adopting digital twins by bringing together drawings, models, GIS, 3D models automatically created from photos and point clouds, and incorporation of disparate data from other enterprise systems like ERP and maintenance management systems as well as real-time IoT data. Rob will share how the professionals involved in design, construction and operations of energy assets can develop and leverage digital twins to improve the efficiency of operations, reduce costs, increase worker safety, and gain new insights to improve asset performance and reliability. One of the key benefits of digital twins is the ability to more effectively visualize and communicate complex project and operations information. But traditional visualization solutions don’t fit well into the new workflows and the potentially very large data sets inherent to digital twins. Bentley’s newly introduced LumenRT for NVIDIA Omniverse showcases how enabling teams to create visualizations directly within digital twin workflows make visualization far more accessible to the average project participant and significantly shorten the time and effort required to create renderings and animations to support better collaboration throughout project delivery and operations and maintenance.
In this session see how Nextspace’s ground-breaking technology unifies IT, OT, and 3D native data, making it accessible and possible to generate interoperable and intelligent Industrial Metaverse applications with Nvidia’s Omniverse Platform
The acts of vandalism last December at Duke Energy’s substations affected 45,000 customers in Moore County, NC, leaving people literally in the dark, with curfews in place and schools unable to function. The need to monitor and protect critical infrastructures that provide energy to millions of people has never been more pressing. This presentation will demonstrate with specific project examples how IronYun’s Vaidio® AI Vision Platform can be uniquely applied to improve employee safety and facility security, increase monitoring efficiency, and potentially increase uptime. We’ll show how Vaidio solutions help strengthen critical utility infrastructure and effectively scale operations.
The pressures of today’s business environment have created new opportunities for industrial innovation. In this session, leaders from Deloitte’s Enterprise Simulation offering will share their perspectives on how organizations might think about their digital twin programs. We will cover the advancements that frame the new “tech stack” for industry and also explore specific examples of putting AI in the center of these new capabilities.
We will discuss the unique approach NVIDIA is taking to help Fortune 500s demystify AI from the earliest stages of App Development to deploying securely at Fortune 5 scale.
The sum of its parts is more than a box. DGX is a full-stack purpose-built & optimized platform for AI workloads. DGX provides maximum performance & scalability from 10s to 1000s of GPUs. Learn about the DGX portfolio and how it can unleash the power of AI within your company.
Physics ML is the term that we use at NVIDIA to talk about solving hard engineering problems using AI: problems like optimizing engineering designs, developing virtual sensors, and building digital twins. Until recently some of these problems were only attempted using sophisticated numerical tools, while others were considered impossible due to the great computational requirements of traditional methods. In this talk, we will discuss how Physics ML is changing what is possible. We’ll share some examples, introduce three different types of Physics ML techniques, the skills required to use them, and discuss what NVIDIA is doing to help reduce the barrier to develop and deploy Physics ML solutions.
The impact of AI for industries such as healthcare, life sciences, and energy is taking shape. Everything from computer vision assisted healthcare diagnostics, AI accelerated protein folding for drug discovery, and audio processing for preventative equipment maintenance can be automated with machine learning.
In this session, you will hear about common challenges ML engineers face in their regular development process and how Run:ai and Weights and Biases provides ML practitioners the tools and resources they need to build better models faster. In this session, we will demonstrate how these tools can help users rapidly train computer vision models, track their results, and build rich reports.
The rise of virtualization, disaggregation, and containerized microservices coupled with software-defined infrastructure has placed an unprecedented level of demand on the modern DC. In addition, enterprise move toward distributed cloud application delivery has given rise to new security threats that simply cannot be ignored. To address these challenges, NVIDIA is leading the charge with a data processing unit (DPU) called BlueField. A necessary component of modern Data Center, BlueField delivers a broad range of advanced networking, security, and storage services for complex compute and AI workloads. Supported by accelerated computing and programmability, BlueField brings an unprecedented level of analytics and security to the edge as well. Regardless of whether you need to improve application performance, implement zero-trust security, leverage composability, or bring security and intelligence to the edge, you will find BlueField an interesting solution to evaluate.
Modern healthcare research, like we’ve seen in other industries, is increasingly being fueled by the combination of data, algorithms, compute, and dedicated bi-lingual domain experts and data science professionals. Deep learning is being used to analyze biomedical imaging, physician notes, omics, labs data and combinations thereof to draw insights beyond what humans can alone. Digital and Technology teams within healthcare organizations must enable new platforms, services, and capabilities to meet the demands of medical researchers to develop and deploy healthcare AI tools. In this talk, we’ll highlight state-of-the-art use-cases of AI in healthcare as well as reference architectures for an “AI Center of Excellence” within a Medical Center to simplify deployment of functional and secure AI platforms, accelerating time to value.
For precision medicine to become routine, genome sequencing needs to be delivered at high accuracy, high speed, low cost, and at scales that drive new understandings in human biology and medicine. GPU-based parallel processing of sequencing data, in addition to accelerated deep learning in base calling and variant calling, is now delivering high performance across the entire genomics computational workflow. This talk will show how NVIDIA is delivering on this end-to-end genomics compute framework, with the latest news on NVIDIA's Clara Parabricks software for genomics analysis. With up to 80x acceleration on popular workflows and high accuracy deep learning approaches, several use cases from critical care rapid sequencing to large-scale population studies are becoming faster and cheaper while getting newer insights.
Artificial Intelligence (AI) is transforming the field of medicine, by providing new opportunities for diagnosing and treating diseases and making surgery safer. Join David Niewolny, NVIDIA’s Business Development Leader for Healthcare/Medical to learn how you can accelerate your AI enabled product to market and increase the pace of innovation through software defined devices. In this presentation, we will explore how AI can aid in accurate and efficient analysis of medical images, and how AI can assist in surgical planning, enable precise navigation, and improve surgical outcomes. The NVIDIA Holoscan SDK and MONAI (Medical Open Network for AI) frameworks will be showcased as key tools that enable the researchers and device developers to efficiently bring to market new AI-powered medical imaging, digital surgery, and robotics products.