AWS re:Invent 2022 Session Schedule

Summary: In this chalk talk, learn about a new physics-informed neural network (PINNs) framework for simulations on PyTorch for fast-to-compute and highly accurate models for various applications. The computational cost associated with multi-physics simulations often impose serious limitations for design space exploration, optimization, and uncertainty quantification. Learn how to apply methods for solving partial differential equations that govern the physics of engineering systems using PINNs and other relevant applications in engineering. This was developed and implemented by AWS in collaboration with the University of Central Florida.

Session Type: Chalk Talk | Session ID: CMP211-R

Summary: File services are often asked to deliver performant capabilities with low latencies for compute-intensive workloads or high IOPS for data-intensive applications. In this session, learn best practices for choosing how to rightsize your performance, reduce latency, increase throughput, and optimize cost. The Amazon FSx family of managed file storage services provides feature sets and performance profiles supporting a variety of use cases. Many workloads like machine learning (ML), high performance computing (HPC), video rendering, and financial analytics require cost optimizations and deployment options. This session also shares tips and best practices for migrating performant-dependent use cases to AWS.

Session Type: Breakout | Session ID: STG343

Summary: In this chalk talk, explore how financial services industry organizations can achieve greater elasticity, higher throughput, and significant cost savings with a cloud-native high-throughput computing (HTC) grid solution built on AWS core services. Using modular architecture made up of managed services such as Amazon EKS, Amazon DynamoDB, Amazon SQS, and Amazon EC2 Spot Instances, HTC grid solutions dynamically scale computing resources to meet the demanding throughput requirements of risk-analytics workloads. Join the talk to learn about the modular architecture, scalability, and performance you can achieve with this solution.

Session Type: Chalk Talk | Session ID: CMP331-R

Summary: In this interactive builders’ session, the leading AWS Batch product manager and team answer common questions about the service. Get the opportunity to provide feedback to the people planning the next set of AWS Batch features. Come prepared to share your questions and thoughts. You must bring your laptop to participate.

Session Type: Builders | Session ID: CMP317-R

Summary: AWS offers many solutions to design, simulate, and verify the advanced semiconductor devices that are the foundation of modern technology. Electronic design automation (EDA) workloads are critical to the success of chip development. EDA requires computing for digital and analog simulations, logic synthesis, design rule checks, and physical verification. In this session, discover best practices for running HPC or EDA workloads on AWS using Amazon EC2. Hear from Arm about how they use AWS to accelerate EDA workloads in the cloud using Arm-based AWS Graviton instances. Also, Marvell shares how they’re using EDA in the cloud to scale their highly differentiated cloud-optimized silicon solutions for customers like AWS.

Session Type: Breakout | Session ID: CMP320

Summary: In this chalk talk, learn about a new physics-informed neural network (PINNs) framework for simulations on PyTorch for fast-to-compute and highly accurate models for various applications. The computational cost associated with multi-physics simulations often impose serious limitations for design space exploration, optimization, and uncertainty quantification. Learn how to apply methods for solving partial differential equations that govern the physics of engineering systems using PINNs and other relevant applications in engineering. This was developed and implemented by AWS in collaboration with the University of Central Florida.

Session Type: Chalk Talk | Session ID: CMP211-R1

Summary: For decades, industries have been using high-performance computing (HPC) to solve problems that are not solvable on a single machine or to shorten problem-solving run time. Traditionally, HPC workloads are launched and managed on the master node, which represents a single point of failure. Building HPC clusters on AWS allows you to keep the same user experience of launching, monitoring, and managing an HPC workload. In this chalk talk, learn how to launch, monitor, and manage an HPC workload using native AWS services, following a seismic processing problem as an example.

Session Type: Chalk Talk | Session ID: ENU402

Summary: Discover the power of combining traditional HPC simulations with AI/ML services to activate new workflows and scientific breakthroughs. In this chalk talk, learn about combined AWS HPC and ML solutions that have evolved to state-of-the-art supercomputing models, powered by AWS services like Elastic Fabric Adapter, AWS ParallelCluster, Amazon EC2 instances, Amazon FSx for Lustre, and AWS Batch. Learn how AWS trained and deployed a new open-source PyTorch-based protein folding model called OpenFold Single Sequence to accelerate drug discovery on AWS using HPC-grade distributed systems. Dive deep on technology innovations and architecture optimizations for HPC-scale ML use cases.

Session Type: Chalk Talk | Session ID: CMP326

Summary: In this session, learn how NVIDIA enables high-performance computing (HPC) on over 200 Amazon EC2 instance types. NVIDIA provides a multi-platform, standards-compliant, vendor-supported solution for HPC application development that supports all major programming models. Its proven compilers, libraries, and software tools support AWS Graviton to maximize developer productivity, enable hardware platform choice, and facilitate an optimal price-performance solution for HPC applications in the cloud. This presentation is brought to you by NVIDIA, an AWS Partner.

Session Type: Breakout | Session ID: PRT220

Summary: In this chalk talk, learn about unexpected developments in broadcast video production that allowed major studio production to move to high performance computing (HPC) on AWS. As part of the solution deep dive, get a refresher on the AWS Nitro System. Learn about Elastic Fabric Adapter (EFA), a network for Amazon EC2 instances and how they enable scalable AI/ML model training for advanced HPC workloads that go beyond broadcasting to drug discovery, weather, and other use cases.

Session Type: Chalk Talk | Session ID: CMP316

Summary: Energy and utility companies use high performance computing (HPC) to design and optimize operations. Elastic, cloud-based HPC resources can dynamically provide the computational power needed for simulations on demand by creating reusable pipelines. HPC on AWS can be instantiated with the nodes and memory to optimize turnaround time while keeping the results sufficiently accurate. This allows you to make decisions about tradeoffs between cost, run time, and accuracy resolution based on your organization’s priorities. This chalk talk explores how to create HPC using infrastructure as code with AWS CloudFormation, Amazon EC2 Image Builder, Amazon FSx for Lustre, and inherited IAM permissions.

Session Type: Chalk Talk | Session ID: ENU401

Summary: In this session, you'll gain an understanding for the recently added AWS Batch support for managing and scheduling large-scale data analysis using Amazon Elastic Kubernetes Service (EKS). Join this hands-on workshop and learn AWS Batch fundamental concepts, and how it works together with your other AWS services running within EKS. You will deploy an EKS cluster, leverage AWS Batch to managed pods, and train a simple machine learning model. You must bring your laptop to participate in hands-on exercises.

Session Type: Workshop | Session ID: CMP335-R

Summary: This talk is delivered in Italian. High performance computing (HPC) has always been about utilizing the most effective technologies to solve the world’s most complex problems. However, many HPC users have been constrained by the need to purchase and deploy on-premises infrastructure to gain access to the best technologies for their applications. AWS HPC helps you innovate beyond the limitations of on-premises HPC infrastructure. In this talk, see how to create and manage HPC environments with AWS ParallelCluster, access 400 Gbps networking with Elastic Fabric Adapter, and chose from over 500 Amazon EC2 instances to run your largest, most complex HPC workloads in the cloud.

Session Type: Geo Talk | Session ID: GEO032

Summary: This chalk talk details best practices and lessons learned about coding for multiple architectures (x86 and AWS Graviton) from the last ten years in the high performance computing (HPC) community. Amazon EC2 has seen a significant rise in the adoption of Arm-based AWS Graviton instances. When supporting multiple architectures, it’s important to ensure a consistent developer experience. When workloads run on multiple architectures, platform nuances and commonalities must be addressed to forge a transparent developer environment. Come learn how HPC development techniques can be applied to almost any compute workload and how to effectively code for multiple compute infrastructure architectures.

Session Type: Chalk Talk | Session ID: CMP325

Summary: Organizations are using Kubernetes to manage their microservices, and it’s natural to want to use it for everything else, including the batch-type workloads in HPC, AI/ML, and large-scale data processing. Customer feedback led AWS to connect AWS Batch to Amazon EKS. The result is a fully managed service for scaling Kubernetes nodes, placing pods, and supporting job completion— the whole lifecycle. In this session, learn how to easily run batch jobs at scale.

Session Type: Demo | Session ID: CMP003

Summary: Join this workshop to gain a solid skills foundation to run common high performance computing (HPC) workloads using cloud technologies. Start with a primer on modern cloud foundations and how they map to common HPC concepts. Then, get a deeper view into how to use cloud-native technologies, such as object storage, in HPC workflows. Get hands-on experience setting up your own cluster in the cloud and running a sample application. You must bring your laptop to participate.

Session Type: Workshop | Session ID: CMP305

Summary: Imagine yourself as an aerodynamics engineer who uses computational fluid dynamics (CFD) to design new cars and planes, and you want the fastest turnaround times while retaining accuracy. This interactive chalk talk walks through techniques for optimizing complex workflows by combining AWS capabilities with optimization tricks to reduce simulation time for popular CFD codes such as OpenFOAM and Siemens Simcenter STAR-CCM+.

Session Type: Chalk Talk | Session ID: CMP402

Summary: In this session, you'll gain an understanding for the recently added AWS Batch support for managing and scheduling large-scale data analysis using Amazon Elastic Kubernetes Service (EKS). Join this hands-on workshop and learn AWS Batch fundamental concepts, and how it works together with your other AWS services running within EKS. You will deploy an EKS cluster, leverage AWS Batch to managed pods, and train a simple machine learning model. You must bring your laptop to participate in hands-on exercises.

Session Type: Workshop | Session ID: CMP335-R1

Summary: AWS and Intel are committed to solving the world’s most critical social and environmental challenges through the power of HPC. Intel and AWS have collaborated to deliver first-rate engineering, resources, and results to impact customers around the world. Join this session to hear from Harvard Cancer Center, IKTOS, and Good Chemistry Co., who have utilized the scalability and availability of cloud computing powered by Intel to address problems spanning from cancer to ‘forever chemicals’ in water. This presentation is brought to you by Intel, an AWS Partner.

Session Type: Breakout | Session ID: PRT278

Summary: In this session, learn how high performance computing (HPC) on AWS makes extreme-scale compute possible so that you can solve some of the world’s toughest environmental, social, health, and scientific problems. Learn about infrastructure that can help you design your products faster with simulations, predict the weather, detect seismic activity with greater precision, and much more. Discover how to run your massively parallel workloads and analyze more data with virtually unlimited infrastructure and nearly endless scale. Finally, learn how you can simply turn it off when you’re done and start it up when you’re ready to begin again, helping you optimize your HPC costs.

Session Type: Breakout | Session ID: CMP205

Summary: In this session, learn how high performance computing (HPC) on AWS makes extreme-scale compute possible so that you can solve some of the world’s toughest environmental, social, health, and scientific problems. Learn about infrastructure that can help you design your products faster with simulations, predict the weather, detect seismic activity with greater precision, and much more. Discover how to run your massively parallel workloads and analyze more data with virtually unlimited infrastructure and nearly endless scale. Finally, learn how you can simply turn it off when you’re done and start it up when you’re ready to begin again, helping you optimize your HPC costs.

Session Type: Breakout | Session ID: CMP205-OF

Summary: In this chalk talk, learn how to efficiently and cost-effectively run your high-scale, compute-intensive workloads using the capabilities of AWS Batch. Scaling containers for batch workloads like machine learning training, simulations, and large dataset processing in a timely fashion is challenging. AWS Batch is a fully managed service for orchestrating AWS compute resources for batch processing or parallel computing at any scale. AWS Batch can schedule, orchestrate, and run your jobs on container-based clusters and automatically scale resources to match your job requirements.

Session Type: Chalk Talk | Session ID: CMP329-R

Summary: As the world strives to solve increasingly complex problems, single on-premises general-purpose supercomputers can fall short. In this session, you’ll learn how AWS is reimagining supercomputing to provide scientists and engineers more access to world-class facilities and technology. Whether it’s finding new life-saving drugs, localizing natural disaster prediction, or designing airplanes powered by alternative energy sources, High Performance Computing (HPC) is transforming our lives. Nearly everything we use, such as our cars and electronics, are designed using supercomputers, a class of HPC computing. Supercomputers divide a single problem across hundreds to thousands of servers to perform high-speed calculations that reduce overall time to solution.

Session Type: Breakout | Session ID: CMP222

Summary: Elastic Fabric Adapter (EFA), built on the Scalable Reliable Datagram (SRD) protocol, is the foundation for scaling message passing interface (MPI) and NVIDIA Collective Communications Library–based high performance computing (HPC) and machine learning (ML) codes at AWS. This builders’ session addresses application scaling challenges and is led by an AWS EFA team principal engineer. It includes a roundtable-style discussion, so come prepared to share your thoughts and questions. You must bring your laptop to participate.

Session Type: Builders | Session ID: CMP407

Summary: New methods like cryo-EM are enabling rapid advances in drug discovery. Using the right analysis workflow and tools are critical to success. In this lightning talk, join WEKA and Clovertex to learn how you can accelerate your drug design workflows using the WEKA data platform and Clovertex’s services on AWS. This presentation is brought to you by WEKA, an AWS Partner.

Session Type: Lightning Talk | Session ID: PRT107

Summary: In this interactive builders’ session, the leading AWS Batch product manager and team answer common questions about the service. Get the opportunity to provide feedback to the people planning the next set of AWS Batch features. Come prepared to share your questions and thoughts. You must bring your laptop to participate.

Session Type: Builders | Session ID: CMP317-R1

Summary: In this session, learn about model-based design and the high-performance computing (HPC) this drives. Hear from AWS customers who are rethinking and reworking their HPC workloads on AWS with a focus on innovation, agility, and the use of deep learning in combination with HPC. Additionally, AWS experts present new approaches for the convergence of model-based design, machine learning, data analytics, and HPC with application to aerospace and engineering challenges.

Session Type: Breakout | Session ID: CMP220

Summary: In this session, learn about model-based design and the high-performance computing (HPC) this drives. Hear from AWS customers who are rethinking and reworking their HPC workloads on AWS with a focus on innovation, agility, and the use of deep learning in combination with HPC. Additionally, AWS experts present new approaches for the convergence of model-based design, machine learning, data analytics, and HPC with application to aerospace and engineering challenges.

Session Type: Breakout | Session ID: CMP220-OF

Summary: In this chalk talk, learn how to efficiently and cost-effectively run your high-scale, compute-intensive workloads using the capabilities of AWS Batch. Scaling containers for batch workloads like machine learning training, simulations, and large dataset processing in a timely fashion is challenging. AWS Batch is a fully managed service for orchestrating AWS compute resources for batch processing or parallel computing at any scale. AWS Batch can schedule, orchestrate, and run your jobs on container-based clusters and automatically scale resources to match your job requirements.

Session Type: Chalk Talk | Session ID: CMP329-R1

Summary: In this interactive builders’ session, the leading AWS Batch product manager and team answer common questions about the service. Get the opportunity to provide feedback to the people planning the next set of AWS Batch features. Come prepared to share your questions and thoughts. You must bring your laptop to participate.

Session Type: Builders | Session ID: CMP317-R

Summary: In this session, you'll gain an understanding for the recently added AWS Batch support for managing and scheduling large-scale data analysis using Amazon Elastic Kubernetes Service (EKS). Join this hands-on workshop and learn AWS Batch fundamental concepts, and how it works together with your other AWS services running within EKS. You will deploy an EKS cluster, leverage AWS Batch to managed pods, and train a simple machine learning model. You must bring your laptop to participate in hands-on exercises.

Session Type: Workshop | Session ID: CMP335-R

Summary: This talk is delivered in Italian. High performance computing (HPC) has always been about utilizing the most effective technologies to solve the world’s most complex problems. However, many HPC users have been constrained by the need to purchase and deploy on-premises infrastructure to gain access to the best technologies for their applications. AWS HPC helps you innovate beyond the limitations of on-premises HPC infrastructure. In this talk, see how to create and manage HPC environments with AWS ParallelCluster, access 400 Gbps networking with Elastic Fabric Adapter, and chose from over 500 Amazon EC2 instances to run your largest, most complex HPC workloads in the cloud.

Session Type: Geo Talk | Session ID: GEO032

Summary: Organizations are using Kubernetes to manage their microservices, and it’s natural to want to use it for everything else, including the batch-type workloads in HPC, AI/ML, and large-scale data processing. Customer feedback led AWS to connect AWS Batch to Amazon EKS. The result is a fully managed service for scaling Kubernetes nodes, placing pods, and supporting job completion— the whole lifecycle. In this session, learn how to easily run batch jobs at scale.

Session Type: Demo | Session ID: CMP003

Summary: Join this workshop to gain a solid skills foundation to run common high performance computing (HPC) workloads using cloud technologies. Start with a primer on modern cloud foundations and how they map to common HPC concepts. Then, get a deeper view into how to use cloud-native technologies, such as object storage, in HPC workflows. Get hands-on experience setting up your own cluster in the cloud and running a sample application. You must bring your laptop to participate.

Session Type: Workshop | Session ID: CMP305

Summary: In this session, you'll gain an understanding for the recently added AWS Batch support for managing and scheduling large-scale data analysis using Amazon Elastic Kubernetes Service (EKS). Join this hands-on workshop and learn AWS Batch fundamental concepts, and how it works together with your other AWS services running within EKS. You will deploy an EKS cluster, leverage AWS Batch to managed pods, and train a simple machine learning model. You must bring your laptop to participate in hands-on exercises.

Session Type: Workshop | Session ID: CMP335-R1

Summary: In this session, learn how high performance computing (HPC) on AWS makes extreme-scale compute possible so that you can solve some of the world’s toughest environmental, social, health, and scientific problems. Learn about infrastructure that can help you design your products faster with simulations, predict the weather, detect seismic activity with greater precision, and much more. Discover how to run your massively parallel workloads and analyze more data with virtually unlimited infrastructure and nearly endless scale. Finally, learn how you can simply turn it off when you’re done and start it up when you’re ready to begin again, helping you optimize your HPC costs.

Session Type: Breakout | Session ID: CMP205

Summary: In this session, learn how high performance computing (HPC) on AWS makes extreme-scale compute possible so that you can solve some of the world’s toughest environmental, social, health, and scientific problems. Learn about infrastructure that can help you design your products faster with simulations, predict the weather, detect seismic activity with greater precision, and much more. Discover how to run your massively parallel workloads and analyze more data with virtually unlimited infrastructure and nearly endless scale. Finally, learn how you can simply turn it off when you’re done and start it up when you’re ready to begin again, helping you optimize your HPC costs.

Session Type: Breakout | Session ID: CMP205-OF

Summary: In this chalk talk, learn how to efficiently and cost-effectively run your high-scale, compute-intensive workloads using the capabilities of AWS Batch. Scaling containers for batch workloads like machine learning training, simulations, and large dataset processing in a timely fashion is challenging. AWS Batch is a fully managed service for orchestrating AWS compute resources for batch processing or parallel computing at any scale. AWS Batch can schedule, orchestrate, and run your jobs on container-based clusters and automatically scale resources to match your job requirements.

Session Type: Chalk Talk | Session ID: CMP329-R

Summary: As the world strives to solve increasingly complex problems, single on-premises general-purpose supercomputers can fall short. In this session, you’ll learn how AWS is reimagining supercomputing to provide scientists and engineers more access to world-class facilities and technology. Whether it’s finding new life-saving drugs, localizing natural disaster prediction, or designing airplanes powered by alternative energy sources, High Performance Computing (HPC) is transforming our lives. Nearly everything we use, such as our cars and electronics, are designed using supercomputers, a class of HPC computing. Supercomputers divide a single problem across hundreds to thousands of servers to perform high-speed calculations that reduce overall time to solution.

Session Type: Breakout | Session ID: CMP222

Summary: In this interactive builders’ session, the leading AWS Batch product manager and team answer common questions about the service. Get the opportunity to provide feedback to the people planning the next set of AWS Batch features. Come prepared to share your questions and thoughts. You must bring your laptop to participate.

Session Type: Builders | Session ID: CMP317-R1

Summary: In this session, learn about model-based design and the high-performance computing (HPC) this drives. Hear from AWS customers who are rethinking and reworking their HPC workloads on AWS with a focus on innovation, agility, and the use of deep learning in combination with HPC. Additionally, AWS experts present new approaches for the convergence of model-based design, machine learning, data analytics, and HPC with application to aerospace and engineering challenges.

Session Type: Breakout | Session ID: CMP220

Summary: In this session, learn about model-based design and the high-performance computing (HPC) this drives. Hear from AWS customers who are rethinking and reworking their HPC workloads on AWS with a focus on innovation, agility, and the use of deep learning in combination with HPC. Additionally, AWS experts present new approaches for the convergence of model-based design, machine learning, data analytics, and HPC with application to aerospace and engineering challenges.

Session Type: Breakout | Session ID: CMP220-OF

Summary: In this chalk talk, learn how to efficiently and cost-effectively run your high-scale, compute-intensive workloads using the capabilities of AWS Batch. Scaling containers for batch workloads like machine learning training, simulations, and large dataset processing in a timely fashion is challenging. AWS Batch is a fully managed service for orchestrating AWS compute resources for batch processing or parallel computing at any scale. AWS Batch can schedule, orchestrate, and run your jobs on container-based clusters and automatically scale resources to match your job requirements.

Session Type: Chalk Talk | Session ID: CMP329-R1

Summary: In this session, learn how NVIDIA enables high-performance computing (HPC) on over 200 Amazon EC2 instance types. NVIDIA provides a multi-platform, standards-compliant, vendor-supported solution for HPC application development that supports all major programming models. Its proven compilers, libraries, and software tools support AWS Graviton to maximize developer productivity, enable hardware platform choice, and facilitate an optimal price-performance solution for HPC applications in the cloud. This presentation is brought to you by NVIDIA, an AWS Partner.

Session Type: Breakout | Session ID: PRT220

Summary: This chalk talk details best practices and lessons learned about coding for multiple architectures (x86 and AWS Graviton) from the last ten years in the high performance computing (HPC) community. Amazon EC2 has seen a significant rise in the adoption of Arm-based AWS Graviton instances. When supporting multiple architectures, it’s important to ensure a consistent developer experience. When workloads run on multiple architectures, platform nuances and commonalities must be addressed to forge a transparent developer environment. Come learn how HPC development techniques can be applied to almost any compute workload and how to effectively code for multiple compute infrastructure architectures.

Session Type: Chalk Talk | Session ID: CMP325

Summary: Imagine yourself as an aerodynamics engineer who uses computational fluid dynamics (CFD) to design new cars and planes, and you want the fastest turnaround times while retaining accuracy. This interactive chalk talk walks through techniques for optimizing complex workflows by combining AWS capabilities with optimization tricks to reduce simulation time for popular CFD codes such as OpenFOAM and Siemens Simcenter STAR-CCM+.

Session Type: Chalk Talk | Session ID: CMP402

Summary: Imagine yourself as an aerodynamics engineer who uses computational fluid dynamics (CFD) to design new cars and planes, and you want the fastest turnaround times while retaining accuracy. This interactive chalk talk walks through techniques for optimizing complex workflows by combining AWS capabilities with optimization tricks to reduce simulation time for popular CFD codes such as OpenFOAM and Siemens Simcenter STAR-CCM+.

Session Type: Chalk Talk | Session ID: CMP402

Summary: In this interactive builders’ session, the leading AWS Batch product manager and team answer common questions about the service. Get the opportunity to provide feedback to the people planning the next set of AWS Batch features. Come prepared to share your questions and thoughts. You must bring your laptop to participate.

Session Type: Builders | Session ID: CMP317-R

Summary: In this session, you'll gain an understanding for the recently added AWS Batch support for managing and scheduling large-scale data analysis using Amazon Elastic Kubernetes Service (EKS). Join this hands-on workshop and learn AWS Batch fundamental concepts, and how it works together with your other AWS services running within EKS. You will deploy an EKS cluster, leverage AWS Batch to managed pods, and train a simple machine learning model. You must bring your laptop to participate in hands-on exercises.

Session Type: Workshop | Session ID: CMP335-R

Summary: Organizations are using Kubernetes to manage their microservices, and it’s natural to want to use it for everything else, including the batch-type workloads in HPC, AI/ML, and large-scale data processing. Customer feedback led AWS to connect AWS Batch to Amazon EKS. The result is a fully managed service for scaling Kubernetes nodes, placing pods, and supporting job completion— the whole lifecycle. In this session, learn how to easily run batch jobs at scale.

Session Type: Demo | Session ID: CMP003

Summary: In this session, you'll gain an understanding for the recently added AWS Batch support for managing and scheduling large-scale data analysis using Amazon Elastic Kubernetes Service (EKS). Join this hands-on workshop and learn AWS Batch fundamental concepts, and how it works together with your other AWS services running within EKS. You will deploy an EKS cluster, leverage AWS Batch to managed pods, and train a simple machine learning model. You must bring your laptop to participate in hands-on exercises.

Session Type: Workshop | Session ID: CMP335-R1

Summary: In this chalk talk, learn how to efficiently and cost-effectively run your high-scale, compute-intensive workloads using the capabilities of AWS Batch. Scaling containers for batch workloads like machine learning training, simulations, and large dataset processing in a timely fashion is challenging. AWS Batch is a fully managed service for orchestrating AWS compute resources for batch processing or parallel computing at any scale. AWS Batch can schedule, orchestrate, and run your jobs on container-based clusters and automatically scale resources to match your job requirements.

Session Type: Chalk Talk | Session ID: CMP329-R

Summary: In this interactive builders’ session, the leading AWS Batch product manager and team answer common questions about the service. Get the opportunity to provide feedback to the people planning the next set of AWS Batch features. Come prepared to share your questions and thoughts. You must bring your laptop to participate.

Session Type: Builders | Session ID: CMP317-R1

Summary: In this chalk talk, learn how to efficiently and cost-effectively run your high-scale, compute-intensive workloads using the capabilities of AWS Batch. Scaling containers for batch workloads like machine learning training, simulations, and large dataset processing in a timely fashion is challenging. AWS Batch is a fully managed service for orchestrating AWS compute resources for batch processing or parallel computing at any scale. AWS Batch can schedule, orchestrate, and run your jobs on container-based clusters and automatically scale resources to match your job requirements.

Session Type: Chalk Talk | Session ID: CMP329-R1

Summary: This chalk talk details best practices and lessons learned about coding for multiple architectures (x86 and AWS Graviton) from the last ten years in the high performance computing (HPC) community. Amazon EC2 has seen a significant rise in the adoption of Arm-based AWS Graviton instances. When supporting multiple architectures, it’s important to ensure a consistent developer experience. When workloads run on multiple architectures, platform nuances and commonalities must be addressed to forge a transparent developer environment. Come learn how HPC development techniques can be applied to almost any compute workload and how to effectively code for multiple compute infrastructure architectures.

Session Type: Chalk Talk | Session ID: CMP325

Summary: AWS offers many solutions to design, simulate, and verify the advanced semiconductor devices that are the foundation of modern technology. Electronic design automation (EDA) workloads are critical to the success of chip development. EDA requires computing for digital and analog simulations, logic synthesis, design rule checks, and physical verification. In this session, discover best practices for running HPC or EDA workloads on AWS using Amazon EC2. Hear from Arm about how they use AWS to accelerate EDA workloads in the cloud using Arm-based AWS Graviton instances. Also, Marvell shares how they’re using EDA in the cloud to scale their highly differentiated cloud-optimized silicon solutions for customers like AWS.

Session Type: Breakout | Session ID: CMP320

Summary: In this chalk talk, learn about a new physics-informed neural network (PINNs) framework for simulations on PyTorch for fast-to-compute and highly accurate models for various applications. The computational cost associated with multi-physics simulations often impose serious limitations for design space exploration, optimization, and uncertainty quantification. Learn how to apply methods for solving partial differential equations that govern the physics of engineering systems using PINNs and other relevant applications in engineering. This was developed and implemented by AWS in collaboration with the University of Central Florida.

Session Type: Chalk Talk | Session ID: CMP211-R

Summary: In this chalk talk, learn about a new physics-informed neural network (PINNs) framework for simulations on PyTorch for fast-to-compute and highly accurate models for various applications. The computational cost associated with multi-physics simulations often impose serious limitations for design space exploration, optimization, and uncertainty quantification. Learn how to apply methods for solving partial differential equations that govern the physics of engineering systems using PINNs and other relevant applications in engineering. This was developed and implemented by AWS in collaboration with the University of Central Florida.

Session Type: Chalk Talk | Session ID: CMP211-R1

Summary: In this chalk talk, learn about unexpected developments in broadcast video production that allowed major studio production to move to high performance computing (HPC) on AWS. As part of the solution deep dive, get a refresher on the AWS Nitro System. Learn about Elastic Fabric Adapter (EFA), a network for Amazon EC2 instances and how they enable scalable AI/ML model training for advanced HPC workloads that go beyond broadcasting to drug discovery, weather, and other use cases.

Session Type: Chalk Talk | Session ID: CMP316

Summary: This talk is delivered in Italian. High performance computing (HPC) has always been about utilizing the most effective technologies to solve the world’s most complex problems. However, many HPC users have been constrained by the need to purchase and deploy on-premises infrastructure to gain access to the best technologies for their applications. AWS HPC helps you innovate beyond the limitations of on-premises HPC infrastructure. In this talk, see how to create and manage HPC environments with AWS ParallelCluster, access 400 Gbps networking with Elastic Fabric Adapter, and chose from over 500 Amazon EC2 instances to run your largest, most complex HPC workloads in the cloud.

Session Type: Geo Talk | Session ID: GEO032

Summary: Join this workshop to gain a solid skills foundation to run common high performance computing (HPC) workloads using cloud technologies. Start with a primer on modern cloud foundations and how they map to common HPC concepts. Then, get a deeper view into how to use cloud-native technologies, such as object storage, in HPC workflows. Get hands-on experience setting up your own cluster in the cloud and running a sample application. You must bring your laptop to participate.

Session Type: Workshop | Session ID: CMP305

Summary: In this session, learn how high performance computing (HPC) on AWS makes extreme-scale compute possible so that you can solve some of the world’s toughest environmental, social, health, and scientific problems. Learn about infrastructure that can help you design your products faster with simulations, predict the weather, detect seismic activity with greater precision, and much more. Discover how to run your massively parallel workloads and analyze more data with virtually unlimited infrastructure and nearly endless scale. Finally, learn how you can simply turn it off when you’re done and start it up when you’re ready to begin again, helping you optimize your HPC costs.

Session Type: Breakout | Session ID: CMP205

Summary: In this session, learn how high performance computing (HPC) on AWS makes extreme-scale compute possible so that you can solve some of the world’s toughest environmental, social, health, and scientific problems. Learn about infrastructure that can help you design your products faster with simulations, predict the weather, detect seismic activity with greater precision, and much more. Discover how to run your massively parallel workloads and analyze more data with virtually unlimited infrastructure and nearly endless scale. Finally, learn how you can simply turn it off when you’re done and start it up when you’re ready to begin again, helping you optimize your HPC costs.

Session Type: Breakout | Session ID: CMP205-OF

Summary: As the world strives to solve increasingly complex problems, single on-premises general-purpose supercomputers can fall short. In this session, you’ll learn how AWS is reimagining supercomputing to provide scientists and engineers more access to world-class facilities and technology. Whether it’s finding new life-saving drugs, localizing natural disaster prediction, or designing airplanes powered by alternative energy sources, High Performance Computing (HPC) is transforming our lives. Nearly everything we use, such as our cars and electronics, are designed using supercomputers, a class of HPC computing. Supercomputers divide a single problem across hundreds to thousands of servers to perform high-speed calculations that reduce overall time to solution.

Session Type: Breakout | Session ID: CMP222

Summary: Elastic Fabric Adapter (EFA), built on the Scalable Reliable Datagram (SRD) protocol, is the foundation for scaling message passing interface (MPI) and NVIDIA Collective Communications Library–based high performance computing (HPC) and machine learning (ML) codes at AWS. This builders’ session addresses application scaling challenges and is led by an AWS EFA team principal engineer. It includes a roundtable-style discussion, so come prepared to share your thoughts and questions. You must bring your laptop to participate.

Session Type: Builders | Session ID: CMP407

Summary: Energy and utility companies use high performance computing (HPC) to design and optimize operations. Elastic, cloud-based HPC resources can dynamically provide the computational power needed for simulations on demand by creating reusable pipelines. HPC on AWS can be instantiated with the nodes and memory to optimize turnaround time while keeping the results sufficiently accurate. This allows you to make decisions about tradeoffs between cost, run time, and accuracy resolution based on your organization’s priorities. This chalk talk explores how to create HPC using infrastructure as code with AWS CloudFormation, Amazon EC2 Image Builder, Amazon FSx for Lustre, and inherited IAM permissions.

Session Type: Chalk Talk | Session ID: ENU401

Summary: In this chalk talk, explore how financial services industry organizations can achieve greater elasticity, higher throughput, and significant cost savings with a cloud-native high-throughput computing (HTC) grid solution built on AWS core services. Using modular architecture made up of managed services such as Amazon EKS, Amazon DynamoDB, Amazon SQS, and Amazon EC2 Spot Instances, HTC grid solutions dynamically scale computing resources to meet the demanding throughput requirements of risk-analytics workloads. Join the talk to learn about the modular architecture, scalability, and performance you can achieve with this solution.

Session Type: Chalk Talk | Session ID: CMP331-R

Summary: In this session, learn about model-based design and the high-performance computing (HPC) this drives. Hear from AWS customers who are rethinking and reworking their HPC workloads on AWS with a focus on innovation, agility, and the use of deep learning in combination with HPC. Additionally, AWS experts present new approaches for the convergence of model-based design, machine learning, data analytics, and HPC with application to aerospace and engineering challenges.

Session Type: Breakout | Session ID: CMP220

Summary: In this session, learn about model-based design and the high-performance computing (HPC) this drives. Hear from AWS customers who are rethinking and reworking their HPC workloads on AWS with a focus on innovation, agility, and the use of deep learning in combination with HPC. Additionally, AWS experts present new approaches for the convergence of model-based design, machine learning, data analytics, and HPC with application to aerospace and engineering challenges.

Session Type: Breakout | Session ID: CMP220-OF

Summary: In this session, learn how NVIDIA enables high-performance computing (HPC) on over 200 Amazon EC2 instance types. NVIDIA provides a multi-platform, standards-compliant, vendor-supported solution for HPC application development that supports all major programming models. Its proven compilers, libraries, and software tools support AWS Graviton to maximize developer productivity, enable hardware platform choice, and facilitate an optimal price-performance solution for HPC applications in the cloud. This presentation is brought to you by NVIDIA, an AWS Partner.

Session Type: Breakout | Session ID: PRT220

Summary: This chalk talk details best practices and lessons learned about coding for multiple architectures (x86 and AWS Graviton) from the last ten years in the high performance computing (HPC) community. Amazon EC2 has seen a significant rise in the adoption of Arm-based AWS Graviton instances. When supporting multiple architectures, it’s important to ensure a consistent developer experience. When workloads run on multiple architectures, platform nuances and commonalities must be addressed to forge a transparent developer environment. Come learn how HPC development techniques can be applied to almost any compute workload and how to effectively code for multiple compute infrastructure architectures.

Session Type: Chalk Talk | Session ID: CMP325

Summary: File services are often asked to deliver performant capabilities with low latencies for compute-intensive workloads or high IOPS for data-intensive applications. In this session, learn best practices for choosing how to rightsize your performance, reduce latency, increase throughput, and optimize cost. The Amazon FSx family of managed file storage services provides feature sets and performance profiles supporting a variety of use cases. Many workloads like machine learning (ML), high performance computing (HPC), video rendering, and financial analytics require cost optimizations and deployment options. This session also shares tips and best practices for migrating performant-dependent use cases to AWS.

Session Type: Breakout | Session ID: STG343

Summary: AWS offers many solutions to design, simulate, and verify the advanced semiconductor devices that are the foundation of modern technology. Electronic design automation (EDA) workloads are critical to the success of chip development. EDA requires computing for digital and analog simulations, logic synthesis, design rule checks, and physical verification. In this session, discover best practices for running HPC or EDA workloads on AWS using Amazon EC2. Hear from Arm about how they use AWS to accelerate EDA workloads in the cloud using Arm-based AWS Graviton instances. Also, Marvell shares how they’re using EDA in the cloud to scale their highly differentiated cloud-optimized silicon solutions for customers like AWS.

Session Type: Breakout | Session ID: CMP320

Summary: For decades, industries have been using high-performance computing (HPC) to solve problems that are not solvable on a single machine or to shorten problem-solving run time. Traditionally, HPC workloads are launched and managed on the master node, which represents a single point of failure. Building HPC clusters on AWS allows you to keep the same user experience of launching, monitoring, and managing an HPC workload. In this chalk talk, learn how to launch, monitor, and manage an HPC workload using native AWS services, following a seismic processing problem as an example.

Session Type: Chalk Talk | Session ID: ENU402

Summary: Discover the power of combining traditional HPC simulations with AI/ML services to activate new workflows and scientific breakthroughs. In this chalk talk, learn about combined AWS HPC and ML solutions that have evolved to state-of-the-art supercomputing models, powered by AWS services like Elastic Fabric Adapter, AWS ParallelCluster, Amazon EC2 instances, Amazon FSx for Lustre, and AWS Batch. Learn how AWS trained and deployed a new open-source PyTorch-based protein folding model called OpenFold Single Sequence to accelerate drug discovery on AWS using HPC-grade distributed systems. Dive deep on technology innovations and architecture optimizations for HPC-scale ML use cases.

Session Type: Chalk Talk | Session ID: CMP326

Summary: In this session, learn how NVIDIA enables high-performance computing (HPC) on over 200 Amazon EC2 instance types. NVIDIA provides a multi-platform, standards-compliant, vendor-supported solution for HPC application development that supports all major programming models. Its proven compilers, libraries, and software tools support AWS Graviton to maximize developer productivity, enable hardware platform choice, and facilitate an optimal price-performance solution for HPC applications in the cloud. This presentation is brought to you by NVIDIA, an AWS Partner.

Session Type: Breakout | Session ID: PRT220

Summary: Energy and utility companies use high performance computing (HPC) to design and optimize operations. Elastic, cloud-based HPC resources can dynamically provide the computational power needed for simulations on demand by creating reusable pipelines. HPC on AWS can be instantiated with the nodes and memory to optimize turnaround time while keeping the results sufficiently accurate. This allows you to make decisions about tradeoffs between cost, run time, and accuracy resolution based on your organization’s priorities. This chalk talk explores how to create HPC using infrastructure as code with AWS CloudFormation, Amazon EC2 Image Builder, Amazon FSx for Lustre, and inherited IAM permissions.

Session Type: Chalk Talk | Session ID: ENU401

Summary: This talk is delivered in Italian. High performance computing (HPC) has always been about utilizing the most effective technologies to solve the world’s most complex problems. However, many HPC users have been constrained by the need to purchase and deploy on-premises infrastructure to gain access to the best technologies for their applications. AWS HPC helps you innovate beyond the limitations of on-premises HPC infrastructure. In this talk, see how to create and manage HPC environments with AWS ParallelCluster, access 400 Gbps networking with Elastic Fabric Adapter, and chose from over 500 Amazon EC2 instances to run your largest, most complex HPC workloads in the cloud.

Session Type: Geo Talk | Session ID: GEO032

Summary: Join this workshop to gain a solid skills foundation to run common high performance computing (HPC) workloads using cloud technologies. Start with a primer on modern cloud foundations and how they map to common HPC concepts. Then, get a deeper view into how to use cloud-native technologies, such as object storage, in HPC workflows. Get hands-on experience setting up your own cluster in the cloud and running a sample application. You must bring your laptop to participate.

Session Type: Workshop | Session ID: CMP305

Summary: Imagine yourself as an aerodynamics engineer who uses computational fluid dynamics (CFD) to design new cars and planes, and you want the fastest turnaround times while retaining accuracy. This interactive chalk talk walks through techniques for optimizing complex workflows by combining AWS capabilities with optimization tricks to reduce simulation time for popular CFD codes such as OpenFOAM and Siemens Simcenter STAR-CCM+.

Session Type: Chalk Talk | Session ID: CMP402

Summary: AWS and Intel are committed to solving the world’s most critical social and environmental challenges through the power of HPC. Intel and AWS have collaborated to deliver first-rate engineering, resources, and results to impact customers around the world. Join this session to hear from Harvard Cancer Center, IKTOS, and Good Chemistry Co., who have utilized the scalability and availability of cloud computing powered by Intel to address problems spanning from cancer to ‘forever chemicals’ in water. This presentation is brought to you by Intel, an AWS Partner.

Session Type: Breakout | Session ID: PRT278

Summary: In this session, learn how high performance computing (HPC) on AWS makes extreme-scale compute possible so that you can solve some of the world’s toughest environmental, social, health, and scientific problems. Learn about infrastructure that can help you design your products faster with simulations, predict the weather, detect seismic activity with greater precision, and much more. Discover how to run your massively parallel workloads and analyze more data with virtually unlimited infrastructure and nearly endless scale. Finally, learn how you can simply turn it off when you’re done and start it up when you’re ready to begin again, helping you optimize your HPC costs.

Session Type: Breakout | Session ID: CMP205

Summary: In this session, learn how high performance computing (HPC) on AWS makes extreme-scale compute possible so that you can solve some of the world’s toughest environmental, social, health, and scientific problems. Learn about infrastructure that can help you design your products faster with simulations, predict the weather, detect seismic activity with greater precision, and much more. Discover how to run your massively parallel workloads and analyze more data with virtually unlimited infrastructure and nearly endless scale. Finally, learn how you can simply turn it off when you’re done and start it up when you’re ready to begin again, helping you optimize your HPC costs.

Session Type: Breakout | Session ID: CMP205-OF

Summary: As the world strives to solve increasingly complex problems, single on-premises general-purpose supercomputers can fall short. In this session, you’ll learn how AWS is reimagining supercomputing to provide scientists and engineers more access to world-class facilities and technology. Whether it’s finding new life-saving drugs, localizing natural disaster prediction, or designing airplanes powered by alternative energy sources, High Performance Computing (HPC) is transforming our lives. Nearly everything we use, such as our cars and electronics, are designed using supercomputers, a class of HPC computing. Supercomputers divide a single problem across hundreds to thousands of servers to perform high-speed calculations that reduce overall time to solution.

Session Type: Breakout | Session ID: CMP222

Summary: New methods like cryo-EM are enabling rapid advances in drug discovery. Using the right analysis workflow and tools are critical to success. In this lightning talk, join WEKA and Clovertex to learn how you can accelerate your drug design workflows using the WEKA data platform and Clovertex’s services on AWS. This presentation is brought to you by WEKA, an AWS Partner.

Session Type: Lightning Talk | Session ID: PRT107

Summary: In this session, learn about model-based design and the high-performance computing (HPC) this drives. Hear from AWS customers who are rethinking and reworking their HPC workloads on AWS with a focus on innovation, agility, and the use of deep learning in combination with HPC. Additionally, AWS experts present new approaches for the convergence of model-based design, machine learning, data analytics, and HPC with application to aerospace and engineering challenges.

Session Type: Breakout | Session ID: CMP220

Summary: In this session, learn about model-based design and the high-performance computing (HPC) this drives. Hear from AWS customers who are rethinking and reworking their HPC workloads on AWS with a focus on innovation, agility, and the use of deep learning in combination with HPC. Additionally, AWS experts present new approaches for the convergence of model-based design, machine learning, data analytics, and HPC with application to aerospace and engineering challenges.

Session Type: Breakout | Session ID: CMP220-OF

Summary: In this chalk talk, explore how financial services industry organizations can achieve greater elasticity, higher throughput, and significant cost savings with a cloud-native high-throughput computing (HTC) grid solution built on AWS core services. Using modular architecture made up of managed services such as Amazon EKS, Amazon DynamoDB, Amazon SQS, and Amazon EC2 Spot Instances, HTC grid solutions dynamically scale computing resources to meet the demanding throughput requirements of risk-analytics workloads. Join the talk to learn about the modular architecture, scalability, and performance you can achieve with this solution.

Session Type: Chalk Talk | Session ID: CMP331-R

Summary: AWS and Intel are committed to solving the world’s most critical social and environmental challenges through the power of HPC. Intel and AWS have collaborated to deliver first-rate engineering, resources, and results to impact customers around the world. Join this session to hear from Harvard Cancer Center, IKTOS, and Good Chemistry Co., who have utilized the scalability and availability of cloud computing powered by Intel to address problems spanning from cancer to ‘forever chemicals’ in water. This presentation is brought to you by Intel, an AWS Partner.

Session Type: Breakout | Session ID: PRT278

Summary: In this interactive builders’ session, the leading AWS Batch product manager and team answer common questions about the service. Get the opportunity to provide feedback to the people planning the next set of AWS Batch features. Come prepared to share your questions and thoughts. You must bring your laptop to participate.

Session Type: Builders | Session ID: CMP317-R

Summary: In this session, you'll gain an understanding for the recently added AWS Batch support for managing and scheduling large-scale data analysis using Amazon Elastic Kubernetes Service (EKS). Join this hands-on workshop and learn AWS Batch fundamental concepts, and how it works together with your other AWS services running within EKS. You will deploy an EKS cluster, leverage AWS Batch to managed pods, and train a simple machine learning model. You must bring your laptop to participate in hands-on exercises.

Session Type: Workshop | Session ID: CMP335-R

Summary: Organizations are using Kubernetes to manage their microservices, and it’s natural to want to use it for everything else, including the batch-type workloads in HPC, AI/ML, and large-scale data processing. Customer feedback led AWS to connect AWS Batch to Amazon EKS. The result is a fully managed service for scaling Kubernetes nodes, placing pods, and supporting job completion— the whole lifecycle. In this session, learn how to easily run batch jobs at scale.

Session Type: Demo | Session ID: CMP003

Summary: In this session, you'll gain an understanding for the recently added AWS Batch support for managing and scheduling large-scale data analysis using Amazon Elastic Kubernetes Service (EKS). Join this hands-on workshop and learn AWS Batch fundamental concepts, and how it works together with your other AWS services running within EKS. You will deploy an EKS cluster, leverage AWS Batch to managed pods, and train a simple machine learning model. You must bring your laptop to participate in hands-on exercises.

Session Type: Workshop | Session ID: CMP335-R1

Summary: In this chalk talk, learn how to efficiently and cost-effectively run your high-scale, compute-intensive workloads using the capabilities of AWS Batch. Scaling containers for batch workloads like machine learning training, simulations, and large dataset processing in a timely fashion is challenging. AWS Batch is a fully managed service for orchestrating AWS compute resources for batch processing or parallel computing at any scale. AWS Batch can schedule, orchestrate, and run your jobs on container-based clusters and automatically scale resources to match your job requirements.

Session Type: Chalk Talk | Session ID: CMP329-R

Summary: In this interactive builders’ session, the leading AWS Batch product manager and team answer common questions about the service. Get the opportunity to provide feedback to the people planning the next set of AWS Batch features. Come prepared to share your questions and thoughts. You must bring your laptop to participate.

Session Type: Builders | Session ID: CMP317-R1

Summary: In this chalk talk, learn how to efficiently and cost-effectively run your high-scale, compute-intensive workloads using the capabilities of AWS Batch. Scaling containers for batch workloads like machine learning training, simulations, and large dataset processing in a timely fashion is challenging. AWS Batch is a fully managed service for orchestrating AWS compute resources for batch processing or parallel computing at any scale. AWS Batch can schedule, orchestrate, and run your jobs on container-based clusters and automatically scale resources to match your job requirements.

Session Type: Chalk Talk | Session ID: CMP329-R1

Summary: File services are often asked to deliver performant capabilities with low latencies for compute-intensive workloads or high IOPS for data-intensive applications. In this session, learn best practices for choosing how to rightsize your performance, reduce latency, increase throughput, and optimize cost. The Amazon FSx family of managed file storage services provides feature sets and performance profiles supporting a variety of use cases. Many workloads like machine learning (ML), high performance computing (HPC), video rendering, and financial analytics require cost optimizations and deployment options. This session also shares tips and best practices for migrating performant-dependent use cases to AWS.

Session Type: Breakout | Session ID: STG343

Summary: In this chalk talk, learn about a new physics-informed neural network (PINNs) framework for simulations on PyTorch for fast-to-compute and highly accurate models for various applications. The computational cost associated with multi-physics simulations often impose serious limitations for design space exploration, optimization, and uncertainty quantification. Learn how to apply methods for solving partial differential equations that govern the physics of engineering systems using PINNs and other relevant applications in engineering. This was developed and implemented by AWS in collaboration with the University of Central Florida.

Session Type: Chalk Talk | Session ID: CMP211-R

Summary: File services are often asked to deliver performant capabilities with low latencies for compute-intensive workloads or high IOPS for data-intensive applications. In this session, learn best practices for choosing how to rightsize your performance, reduce latency, increase throughput, and optimize cost. The Amazon FSx family of managed file storage services provides feature sets and performance profiles supporting a variety of use cases. Many workloads like machine learning (ML), high performance computing (HPC), video rendering, and financial analytics require cost optimizations and deployment options. This session also shares tips and best practices for migrating performant-dependent use cases to AWS.

Session Type: Breakout | Session ID: STG343

Summary: In this chalk talk, learn about a new physics-informed neural network (PINNs) framework for simulations on PyTorch for fast-to-compute and highly accurate models for various applications. The computational cost associated with multi-physics simulations often impose serious limitations for design space exploration, optimization, and uncertainty quantification. Learn how to apply methods for solving partial differential equations that govern the physics of engineering systems using PINNs and other relevant applications in engineering. This was developed and implemented by AWS in collaboration with the University of Central Florida.

Session Type: Chalk Talk | Session ID: CMP211-R1

Summary: Discover the power of combining traditional HPC simulations with AI/ML services to activate new workflows and scientific breakthroughs. In this chalk talk, learn about combined AWS HPC and ML solutions that have evolved to state-of-the-art supercomputing models, powered by AWS services like Elastic Fabric Adapter, AWS ParallelCluster, Amazon EC2 instances, Amazon FSx for Lustre, and AWS Batch. Learn how AWS trained and deployed a new open-source PyTorch-based protein folding model called OpenFold Single Sequence to accelerate drug discovery on AWS using HPC-grade distributed systems. Dive deep on technology innovations and architecture optimizations for HPC-scale ML use cases.

Session Type: Chalk Talk | Session ID: CMP326

Summary: Elastic Fabric Adapter (EFA), built on the Scalable Reliable Datagram (SRD) protocol, is the foundation for scaling message passing interface (MPI) and NVIDIA Collective Communications Library–based high performance computing (HPC) and machine learning (ML) codes at AWS. This builders’ session addresses application scaling challenges and is led by an AWS EFA team principal engineer. It includes a roundtable-style discussion, so come prepared to share your thoughts and questions. You must bring your laptop to participate.

Session Type: Builders | Session ID: CMP407

Summary: In this chalk talk, learn about unexpected developments in broadcast video production that allowed major studio production to move to high performance computing (HPC) on AWS. As part of the solution deep dive, get a refresher on the AWS Nitro System. Learn about Elastic Fabric Adapter (EFA), a network for Amazon EC2 instances and how they enable scalable AI/ML model training for advanced HPC workloads that go beyond broadcasting to drug discovery, weather, and other use cases.

Session Type: Chalk Talk | Session ID: CMP316

Summary: In this chalk talk, learn about unexpected developments in broadcast video production that allowed major studio production to move to high performance computing (HPC) on AWS. As part of the solution deep dive, get a refresher on the AWS Nitro System. Learn about Elastic Fabric Adapter (EFA), a network for Amazon EC2 instances and how they enable scalable AI/ML model training for advanced HPC workloads that go beyond broadcasting to drug discovery, weather, and other use cases.

Session Type: Chalk Talk | Session ID: CMP316

Summary: In this session, learn about model-based design and the high-performance computing (HPC) this drives. Hear from AWS customers who are rethinking and reworking their HPC workloads on AWS with a focus on innovation, agility, and the use of deep learning in combination with HPC. Additionally, AWS experts present new approaches for the convergence of model-based design, machine learning, data analytics, and HPC with application to aerospace and engineering challenges.

Session Type: Breakout | Session ID: CMP220

Summary: In this session, learn about model-based design and the high-performance computing (HPC) this drives. Hear from AWS customers who are rethinking and reworking their HPC workloads on AWS with a focus on innovation, agility, and the use of deep learning in combination with HPC. Additionally, AWS experts present new approaches for the convergence of model-based design, machine learning, data analytics, and HPC with application to aerospace and engineering challenges.

Session Type: Breakout | Session ID: CMP220-OF

Summary: This talk is delivered in Italian. High performance computing (HPC) has always been about utilizing the most effective technologies to solve the world’s most complex problems. However, many HPC users have been constrained by the need to purchase and deploy on-premises infrastructure to gain access to the best technologies for their applications. AWS HPC helps you innovate beyond the limitations of on-premises HPC infrastructure. In this talk, see how to create and manage HPC environments with AWS ParallelCluster, access 400 Gbps networking with Elastic Fabric Adapter, and chose from over 500 Amazon EC2 instances to run your largest, most complex HPC workloads in the cloud.

Session Type: Geo Talk | Session ID: GEO032

Summary: Join this workshop to gain a solid skills foundation to run common high performance computing (HPC) workloads using cloud technologies. Start with a primer on modern cloud foundations and how they map to common HPC concepts. Then, get a deeper view into how to use cloud-native technologies, such as object storage, in HPC workflows. Get hands-on experience setting up your own cluster in the cloud and running a sample application. You must bring your laptop to participate.

Session Type: Workshop | Session ID: CMP305

Summary: In this session, learn how high performance computing (HPC) on AWS makes extreme-scale compute possible so that you can solve some of the world’s toughest environmental, social, health, and scientific problems. Learn about infrastructure that can help you design your products faster with simulations, predict the weather, detect seismic activity with greater precision, and much more. Discover how to run your massively parallel workloads and analyze more data with virtually unlimited infrastructure and nearly endless scale. Finally, learn how you can simply turn it off when you’re done and start it up when you’re ready to begin again, helping you optimize your HPC costs.

Session Type: Breakout | Session ID: CMP205

Summary: In this session, learn how high performance computing (HPC) on AWS makes extreme-scale compute possible so that you can solve some of the world’s toughest environmental, social, health, and scientific problems. Learn about infrastructure that can help you design your products faster with simulations, predict the weather, detect seismic activity with greater precision, and much more. Discover how to run your massively parallel workloads and analyze more data with virtually unlimited infrastructure and nearly endless scale. Finally, learn how you can simply turn it off when you’re done and start it up when you’re ready to begin again, helping you optimize your HPC costs.

Session Type: Breakout | Session ID: CMP205-OF

Summary: As the world strives to solve increasingly complex problems, single on-premises general-purpose supercomputers can fall short. In this session, you’ll learn how AWS is reimagining supercomputing to provide scientists and engineers more access to world-class facilities and technology. Whether it’s finding new life-saving drugs, localizing natural disaster prediction, or designing airplanes powered by alternative energy sources, High Performance Computing (HPC) is transforming our lives. Nearly everything we use, such as our cars and electronics, are designed using supercomputers, a class of HPC computing. Supercomputers divide a single problem across hundreds to thousands of servers to perform high-speed calculations that reduce overall time to solution.

Session Type: Breakout | Session ID: CMP222

Summary: Elastic Fabric Adapter (EFA), built on the Scalable Reliable Datagram (SRD) protocol, is the foundation for scaling message passing interface (MPI) and NVIDIA Collective Communications Library–based high performance computing (HPC) and machine learning (ML) codes at AWS. This builders’ session addresses application scaling challenges and is led by an AWS EFA team principal engineer. It includes a roundtable-style discussion, so come prepared to share your thoughts and questions. You must bring your laptop to participate.

Session Type: Builders | Session ID: CMP407

Summary: In this session, learn about model-based design and the high-performance computing (HPC) this drives. Hear from AWS customers who are rethinking and reworking their HPC workloads on AWS with a focus on innovation, agility, and the use of deep learning in combination with HPC. Additionally, AWS experts present new approaches for the convergence of model-based design, machine learning, data analytics, and HPC with application to aerospace and engineering challenges.

Session Type: Breakout | Session ID: CMP220

Summary: In this session, learn about model-based design and the high-performance computing (HPC) this drives. Hear from AWS customers who are rethinking and reworking their HPC workloads on AWS with a focus on innovation, agility, and the use of deep learning in combination with HPC. Additionally, AWS experts present new approaches for the convergence of model-based design, machine learning, data analytics, and HPC with application to aerospace and engineering challenges.

Session Type: Breakout | Session ID: CMP220-OF

Summary: File services are often asked to deliver performant capabilities with low latencies for compute-intensive workloads or high IOPS for data-intensive applications. In this session, learn best practices for choosing how to rightsize your performance, reduce latency, increase throughput, and optimize cost. The Amazon FSx family of managed file storage services provides feature sets and performance profiles supporting a variety of use cases. Many workloads like machine learning (ML), high performance computing (HPC), video rendering, and financial analytics require cost optimizations and deployment options. This session also shares tips and best practices for migrating performant-dependent use cases to AWS.

Session Type: Breakout | Session ID: STG343

Summary: New methods like cryo-EM are enabling rapid advances in drug discovery. Using the right analysis workflow and tools are critical to success. In this lightning talk, join WEKA and Clovertex to learn how you can accelerate your drug design workflows using the WEKA data platform and Clovertex’s services on AWS. This presentation is brought to you by WEKA, an AWS Partner.

Session Type: Lightning Talk | Session ID: PRT107