SC22 AWS Booth #2425 Theater and Demo Schedule

Summary: Run hundreds of thousands of batch computing workloads without managing infrastructure or a job scheduler with AWS Batch. Stop by for live demos including "Introducing AWS Batch Fair Share Scheduling", "Introducing AWS Batch on EKS", "A cloud-native SciOps pipeline for GROMACS", "Run genomics workflows using the Amazon Genomics CLI and AWS Batch"

Summary: Run hundreds of thousands of batch computing workloads without managing infrastructure or a job scheduler with AWS Batch. Stop by for live demos including "Introducing AWS Batch Fair Share Scheduling", "Introducing AWS Batch on EKS", "A cloud-native SciOps pipeline for GROMACS", "Run genomics workflows using the Amazon Genomics CLI and AWS Batch"

Summary: Run hundreds of thousands of batch computing workloads without managing infrastructure or a job scheduler with AWS Batch. Stop by for live demos including "Introducing AWS Batch Fair Share Scheduling", "Introducing AWS Batch on EKS", "A cloud-native SciOps pipeline for GROMACS", "Run genomics workflows using the Amazon Genomics CLI and AWS Batch"

Summary: Did you know that the Amazon Genomics CLI supports the top 4 worklfow languages for genomics? AGC stands up the AWS infrastructure you need for running workflows written in WDL, Nextflow, CWL, or SnakeMake. In this demo, we will show you how to leverage AGC to run your workflows on top of AWS Batch.

Summary: Whether you are training large machine learning models or running genomic sequencing, EDA, video encoding, big data processing, or autonomous vehicle simulation workloads in the cloud, you’re going to need cloud-native scheduling and orchestration. In this session, learn about AWS Batch, a container-centric, fully managed service for job schedule and AWS compute resource orchestration. Dive deep into AWS Batch major components, such as job definition, job queues, job states, compute environments, instance selection, job throughput, and allocation strategies. Also, discover best practices that can help you build optimized AWS Batch architectures and run AWS Batch at scale.

Summary: Run hundreds of thousands of batch computing workloads without managing infrastructure or a job scheduler with AWS Batch. Stop by for live demos including "Introducing AWS Batch Fair Share Scheduling", "Introducing AWS Batch on EKS", "A cloud-native SciOps pipeline for GROMACS", "Run genomics workflows using the Amazon Genomics CLI and AWS Batch"

Summary: The classification of workloads into precise execution patterns is difficult and can lead to suboptimal execution in order to avoid the proliferation of applications, platforms, and techniques. What if you could choose a unified and intelligent workload submission platform to help you take advtange of the speed and scalability of the AWS Cloud? Regardless of your workload type, if it is compute-intensive, YellowDog on AWS can help you manage it. YellowDog is a fast, scalable, simple, Spot-friendly extensible scheduler and provisioning platform for hybrid and cloud. Concentrate on the insights and outcomes your workloads generate, and let YellowDog on AWS take care of everything else.

Summary: Run hundreds of thousands of batch computing workloads without managing infrastructure or a job scheduler with AWS Batch. Stop by for live demos including "Introducing AWS Batch Fair Share Scheduling", "Introducing AWS Batch on EKS", "A cloud-native SciOps pipeline for GROMACS", "Run genomics workflows using the Amazon Genomics CLI and AWS Batch"

Summary: In this demo, learn about a new service powered by AWS that Ansys is preparing to launch: Ansys Gateway. Ansys Gateway enables simpler access and deployment for several industry-leading Ansys simulations on AWS. Ansys is also performance-testing several of its products on Amazon EC2 instances.

Summary: Rescale, a leading provider for high-performance computing built for AWS, helps customers utilize AWS services like Amazon S3, Amazon EC2, and NICE DCV for engineering and scientific workloads. In this demo, join Rescale engineer Chris Ramirez to learn how aerospace engineers are accelerating compute-driven innovation with Rescale’s platform, which delivers HPC built for the cloud in collaboration with AWS. Find out how organizations automate full-stack HPC operations in the cloud to optimize the cost performance of computer-aided engineering simulations like CFD and FEA.

Summary: The TotalCAE platform empowers engineers to submit hundreds of applications to AWS in just a few clicks. In this demo, learn how TotalCAE uses AWS services including Elastic Fabric Adaptor (EFA), NICE DCV, and Amazon FSx for Lustre and HPC-specific instances like hpc6a.48xlarge. Look at client case studies utilizing these AWS technologies with TotalCAE to advance the state of the art of CAE simulation.

Summary: In this session, explore how AWS is empowering customers to push the boundaries of computational fluid dynamics (CFD) and helping engineers and designers explore more designs and bring products to market faster. Learn about the move toward GPU-enabled CFD codes, the benefits of cloud elasticity, and the growth of machine learning and deep learning for accelerating CFD simulations.

Summary: In this session, learn about the benefits of running Altair Radioss on Amazon EC2 Hpc6a instances powered by third-generation AMD EPYC processors. For numerous, diverse industries worldwide, Radioss is a versatile and comprehensive high-performance explicit finite element (FE) solver, based on solid and fluid dynamics. It is a standard, go-to tool to accurately predict dynamic, transient-loading effects on structures and products to improve safety and survivability and to design more robust products. Radioss on Hpc6a instances delivers impressive performance on premises. Discover the scaling and performance improvements achieved from Radioss running on Hpc6a instances.

Summary: The TotalCAE platform makes it possible for engineers to submit hundreds of applications to AWS in just a few clicks. Learn how TotalCAE uses AWS services including Elastic Fabric Adapter (EFA), HPC-specific instances such as hpc6a.48xlarge, NICE DCV, and Amazon FSx for Lustre. In this session, TotalCAE discusses client case studies using these AWS technologies with TotalCAE to advance the state of the art in CAE simulation.

Summary: In this demo, learn about a new service powered by AWS that Ansys is preparing to launch: Ansys Gateway. Ansys Gateway enables simpler access and deployment for several industry-leading Ansys simulations on AWS. Ansys is also performance-testing several of its products on Amazon EC2 instances.

Summary: Rescale, a leading provider for high-performance computing built for AWS, helps customers utilize AWS services like Amazon S3, Amazon EC2, and NICE DCV for engineering and scientific workloads. In this demo, join Rescale engineer Chris Ramirez to learn how aerospace engineers are accelerating compute-driven innovation with Rescale’s platform, which delivers HPC built for the cloud in collaboration with AWS. Find out how organizations automate full-stack HPC operations in the cloud to optimize the cost performance of computer-aided engineering simulations like CFD and FEA.

Summary: In this demo, learn about a new service powered by AWS that Ansys is preparing to launch: Ansys Gateway. Ansys Gateway enables simpler access and deployment for several industry-leading Ansys simulations on AWS. Ansys is also performance-testing several of its products on Amazon EC2 instances.

Summary: The TotalCAE platform empowers engineers to submit hundreds of applications to AWS in just a few clicks. In this demo, learn how TotalCAE uses AWS services including Elastic Fabric Adaptor (EFA), NICE DCV, and Amazon FSx for Lustre and HPC-specific instances like hpc6a.48xlarge. Look at client case studies utilizing these AWS technologies with TotalCAE to advance the state of the art of CAE simulation.

Summary: Rescale is a premier provider for high performance computing built for AWS. Organizations across all industries have found immense value in using the AWS full-service platform. Join this session to learn how to quickly and efficiently run your various simulation packages on infrastructure like C6i instances, powered by Intel’s Xeon Ice Lake processor.

Summary: Rescale, a leading provider for high-performance computing built for AWS, helps customers utilize AWS services like Amazon S3, Amazon EC2, and NICE DCV for engineering and scientific workloads. In this demo, join Rescale engineer Chris Ramirez to learn how aerospace engineers are accelerating compute-driven innovation with Rescale’s platform, which delivers HPC built for the cloud in collaboration with AWS. Find out how organizations automate full-stack HPC operations in the cloud to optimize the cost performance of computer-aided engineering simulations like CFD and FEA.

Summary: Graph Analytics is a form of data analysis that enables us to find and understand complex relationships between data entities. In this presentation, we dive into current work on algorithms and methods that build in network latency tolerance to perform large-scale graph calculations, and discuss results using Elastic Fabric Adapter (EFA), the network interface for Amazon EC2 instances targeted to run applications requiring high levels of inter-node communications on AWS.

Summary: Arm Forge—widely used in HPC labs in traditional environments and the cloud—makes HPC debugging and profiling on Amazon EC2 simple regardless of which processor architecture you use. In this demo, learn how Arm Forge offers you tools that provide seamless access to remote servers, and dive deep on profiling experiences with AWS Graviton3 on HPC applications.

Summary: In this talk we deep-dive into the origins and technology behind our specialized HPC & machine learning network interface, EFA.

Summary: In this session, learn about the benefits of running Altair Radioss on Amazon EC2 Hpc6a instances powered by third-generation AMD EPYC processors. For numerous, diverse industries worldwide, Radioss is a versatile and comprehensive high-performance explicit finite element (FE) solver, based on solid and fluid dynamics. It is a standard, go-to tool to accurately predict dynamic, transient-loading effects on structures and products to improve safety and survivability and to design more robust products. Radioss on Hpc6a instances delivers impressive performance on premises. Discover the scaling and performance improvements achieved from Radioss running on Hpc6a instances.

Summary: The TotalCAE platform makes it possible for engineers to submit hundreds of applications to AWS in just a few clicks. Learn how TotalCAE uses AWS services including Elastic Fabric Adapter (EFA), HPC-specific instances such as hpc6a.48xlarge, NICE DCV, and Amazon FSx for Lustre. In this session, TotalCAE discusses client case studies using these AWS technologies with TotalCAE to advance the state of the art in CAE simulation.

Summary: The TotalCAE platform empowers engineers to submit hundreds of applications to AWS in just a few clicks. In this demo, learn how TotalCAE uses AWS services including Elastic Fabric Adaptor (EFA), NICE DCV, and Amazon FSx for Lustre and HPC-specific instances like hpc6a.48xlarge. Look at client case studies utilizing these AWS technologies with TotalCAE to advance the state of the art of CAE simulation.

Summary: This session demonstrates how the NVIDIA HPC SDK activates high performance computing (HPC) on over 200 Amazon EC2 instance types comprised of a variety of Arm or x86 CPUs, with or without GPUs. The NVIDIA HPC SDK is a multi-platform, standards-compliant, vendor-supported solution for HPC application development that supports programming models like OpenMP, OpenACC, and CUDA together with ISO standard parallel features in C++ and Fortran. Its proven compilers, libraries, and software tools support AWS Graviton, x86, POWER, and NVIDIA GPUs to maximize developer productivity, activate hardware platform choice, and facilitate an optimal price-performance solution for HPC applications in the cloud.

Summary: The Arm-based AWS Graviton2 and AWS Graviton3 processors available on AWS offer high core-counts, high memory-bandwidth, and high floating-point performance. This session shares experiences with benchmarking these processors for HPC-style codes. In addition to lower-level testing of memory bandwidth and latency, learn about multi-node performance for larger applications, including benchmarks from SPEChpc. Also, explore programming these Arm-based processors using a diverse range of the latest parallel programming models and many major open-source and vendor compilers, including standard parallelism in C++ and Fortran.

Summary: In this talk we deep-dive into the origins and technology behind our specialized HPC & machine learning network interface, EFA.

Summary: Join this session to discover how Amazon EC2 Hpc6a instances powered by third-generation AMD EPYC processors enable a diverse range of HPC applications, such as CFD, weather forecasting, and molecular dynamics workloads, to scale performantly on AWS. See results from Simcenter STAR-CCM+, Ansys Fluent, WRF, GROMACS, and LAMMPS.

Summary: Rescale is a premier provider for high performance computing built for AWS. Organizations across all industries have found immense value in using the AWS full-service platform. Join this session to learn how to quickly and efficiently run your various simulation packages on infrastructure like C6i instances, powered by Intel’s Xeon Ice Lake processor.

Summary: Rescale, a leading provider for high-performance computing built for AWS, helps customers utilize AWS services like Amazon S3, Amazon EC2, and NICE DCV for engineering and scientific workloads. In this demo, join Rescale engineer Chris Ramirez to learn how aerospace engineers are accelerating compute-driven innovation with Rescale’s platform, which delivers HPC built for the cloud in collaboration with AWS. Find out how organizations automate full-stack HPC operations in the cloud to optimize the cost performance of computer-aided engineering simulations like CFD and FEA.

Summary: Arm Forge—widely used in HPC labs in traditional environments and the cloud—makes HPC debugging and profiling on Amazon EC2 simple regardless of which processor architecture you use. In this demo, learn how Arm Forge offers you tools that provide seamless access to remote servers, and dive deep on profiling experiences with AWS Graviton3 on HPC applications.

Summary: Over the last few years, machine learning (ML), more specifically deep learning (DL), has become important for the HPC community. DL uses frameworks like TensorFlow and PyTorch, which use underlying hardware features for better performance. In this session, learn about how DL frameworks use ML-specific Arm architecture features like BFloat16 for faster inference on AWS Graviton3 compared to other platforms.

Summary: In this talk we deep-dive into the origins and technology behind our specialized HPC & machine learning network interface, EFA.

Summary: Run hundreds of thousands of batch computing workloads without managing infrastructure or a job scheduler with AWS Batch. Stop by for live demos including "Introducing AWS Batch Fair Share Scheduling", "Introducing AWS Batch on EKS", "A cloud-native SciOps pipeline for GROMACS", "Run genomics workflows using the Amazon Genomics CLI and AWS Batch"

Summary: Run hundreds of thousands of batch computing workloads without managing infrastructure or a job scheduler with AWS Batch. Stop by for live demos including "Introducing AWS Batch Fair Share Scheduling", "Introducing AWS Batch on EKS", "A cloud-native SciOps pipeline for GROMACS", "Run genomics workflows using the Amazon Genomics CLI and AWS Batch"

Summary: In this demo, the Atos Nimbix team presents a state-of-the-art container-based solution for multi-site and cloud-based HPC workloads. Learn about this solution, complete with a fully integrated application marketplace featuring curated, ready-to-use click-to-run workflows for simulation applications and AI/ML/DL software.

Summary: Whether you are training large machine learning models or running genomic sequencing, EDA, video encoding, big data processing, or autonomous vehicle simulation workloads in the cloud, you’re going to need cloud-native scheduling and orchestration. In this session, learn about AWS Batch, a container-centric, fully managed service for job schedule and AWS compute resource orchestration. Dive deep into AWS Batch major components, such as job definition, job queues, job states, compute environments, instance selection, job throughput, and allocation strategies. Also, discover best practices that can help you build optimized AWS Batch architectures and run AWS Batch at scale.

Summary: Run hundreds of thousands of batch computing workloads without managing infrastructure or a job scheduler with AWS Batch. Stop by for live demos including "Introducing AWS Batch Fair Share Scheduling", "Introducing AWS Batch on EKS", "A cloud-native SciOps pipeline for GROMACS", "Run genomics workflows using the Amazon Genomics CLI and AWS Batch"

Summary: Run hundreds of thousands of batch computing workloads without managing infrastructure or a job scheduler with AWS Batch. Stop by for live demos including "Introducing AWS Batch Fair Share Scheduling", "Introducing AWS Batch on EKS", "A cloud-native SciOps pipeline for GROMACS", "Run genomics workflows using the Amazon Genomics CLI and AWS Batch"

Summary: Visit our NICE DCV demo station for live demonstrations including "NICE DCV in HPC on AWS", "High-resolution showcase with NICE DCV" and "Streaming High-Performance Cloud Gaming with DCV".

Summary: Visit our NICE DCV demo station for live demonstrations including "NICE DCV in HPC on AWS", "High-resolution showcase with NICE DCV" and "Streaming High-Performance Cloud Gaming with DCV".

Summary: Engineers are using HPC simulations to accelerate the time needed to design and validate new products. To achieve this goal, they need not only to run HPC simulations faster but also to easily modify the input files and visualize the results using their preferred CAD and post-processing software that usually runs on the Windows OS. This demo covers a Windows VDI solution that can create the infrastructure for your users to authenticate and connect to virtual Windows sessions with Amazon AppStream 2.0, submit their HPC jobs to a Linux cluster, and view the output files as they are being written out on an Amazon FSx for NetApp ONTAP file system.

Summary: Engineers are using HPC simulations to accelerate the time needed to design and validate new products. To achieve this goal, they need not only to run HPC simulations faster but also to easily modify the input files and visualize the results using their preferred CAD and post-processing software that usually runs on the Windows OS. This demo covers a Windows VDI solution that can create the infrastructure for your users to authenticate and connect to virtual Windows sessions with Amazon AppStream 2.0, submit their HPC jobs to a Linux cluster, and view the output files as they are being written out on an Amazon FSx for NetApp ONTAP file system.

Summary: Visit our NICE DCV demo station for live demonstrations including "NICE DCV in HPC on AWS", "High-resolution showcase with NICE DCV" and "Streaming High-Performance Cloud Gaming with DCV".

Summary: The TotalCAE platform empowers engineers to submit hundreds of applications to AWS in just a few clicks. In this demo, learn how TotalCAE uses AWS services including Elastic Fabric Adaptor (EFA), NICE DCV, and Amazon FSx for Lustre and HPC-specific instances like hpc6a.48xlarge. Look at client case studies utilizing these AWS technologies with TotalCAE to advance the state of the art of CAE simulation.

Summary: With workloads migrating to the cloud, there is an increasing need to visualize those workloads in a performant manner. In this session, learn about what NICE DCV is and discover all the features that come with it.

Summary: Ray tracing for scientific data provides efficient and realistic lighting and materials for enhanced visual analysis. Recent advances in hardware and software now provide interactive ray tracing on both CPU and GPU platforms. To fully exploit interactive ray tracing in remote computing contexts, researchers require a performant framework that provides an interactive experience even in high latency and low bandwidth contexts. This chalk talk presents how the Texas Advanced Computing Center (TACC) uses NICE DCV and the Intel oneAPI Rendering Toolkit to provide interactive visual analysis.

Summary: When providing access to VDI, it is a best practice to have the remote machines not be public facing. Providing secure access to DCV servers that are not public facing can be achieved with Connection Gateway. In this presentation, we will talk about how Connection Gateway can be used to lock down DCV environments.

Summary: Visit our NICE DCV demo station for live demonstrations including "NICE DCV in HPC on AWS", "High-resolution showcase with NICE DCV" and "Streaming High-Performance Cloud Gaming with DCV".

Summary: In this demo hear from SchedMD, the company behind Slurm workload manager software. Slurm is a free open-source workload manager designed specifically to satisfy the demanding needs of high-performance computing. The software is in widespread use at government laboratories, universities, and companies worldwide and performs workload management for over half of the top 10 systems in the TOP500. Slurm is a scheduler of choice in AWS ParallelCluster.

Summary: Arm Forge—widely used in HPC labs in traditional environments and the cloud—makes HPC debugging and profiling on Amazon EC2 simple regardless of which processor architecture you use. In this demo, learn how Arm Forge offers you tools that provide seamless access to remote servers, and dive deep on profiling experiences with AWS Graviton3 on HPC applications.

Summary: At ISC 2022, Spack and AWS announced Spack’s first public binary cache, which speeds up scientific software deployment by over 20 times. Come to this session for a detailed look at how developers at LLNL and Kitware used AWS and GitLab CI to achieve continuous deployment of signed, trusted binaries from a high-velocity Spack repository. Learn about the challenges and solutions involved in applying both modern DevOps and AWS computing power to an HPC problem space.

Summary: Arm Forge—widely used in HPC labs in traditional environments and the cloud—makes HPC debugging and profiling on Amazon EC2 simple regardless of which processor architecture you use. In this demo, learn how Arm Forge offers you tools that provide seamless access to remote servers, and dive deep on profiling experiences with AWS Graviton3 on HPC applications.

Summary: This session demonstrates how the NVIDIA HPC SDK activates high performance computing (HPC) on over 200 Amazon EC2 instance types comprised of a variety of Arm or x86 CPUs, with or without GPUs. The NVIDIA HPC SDK is a multi-platform, standards-compliant, vendor-supported solution for HPC application development that supports programming models like OpenMP, OpenACC, and CUDA together with ISO standard parallel features in C++ and Fortran. Its proven compilers, libraries, and software tools support AWS Graviton, x86, POWER, and NVIDIA GPUs to maximize developer productivity, activate hardware platform choice, and facilitate an optimal price-performance solution for HPC applications in the cloud.

Summary: Arm Forge—widely used in HPC labs in traditional environments and the cloud—makes HPC debugging and profiling on Amazon EC2 simple regardless of which processor architecture you use. In this demo, learn how Arm Forge offers you tools that provide seamless access to remote servers, and dive deep on profiling experiences with AWS Graviton3 on HPC applications.

Summary: Graph Analytics is a form of data analysis that enables us to find and understand complex relationships between data entities. In this presentation, we dive into current work on algorithms and methods that build in network latency tolerance to perform large-scale graph calculations, and discuss results using Elastic Fabric Adapter (EFA), the network interface for Amazon EC2 instances targeted to run applications requiring high levels of inter-node communications on AWS.

Summary: In this live demo, learn about the Parallel Works single-pane-of-glass, multi-cloud environment, which is operated by General Dynamics Information Technology for NOAA’s R&D HPC organization. The presentation demonstrates key aspects of the implementation and optimization of the platform for large-scale HPC-driven R&D. The Parallel Works platform uses 20+ AWS services, including Amazon FSx for Lustre, Amazon S3, Amazon EC2, and Amazon EFS. Dive deep on the end user experience, which incorporates both the familiar on-premises computing environment and the flexibility and innovation of the cloud.

Summary: In this talk we deep-dive into the origins and technology behind our specialized HPC & machine learning network interface, EFA.

Summary: The TotalCAE platform makes it possible for engineers to submit hundreds of applications to AWS in just a few clicks. Learn how TotalCAE uses AWS services including Elastic Fabric Adapter (EFA), HPC-specific instances such as hpc6a.48xlarge, NICE DCV, and Amazon FSx for Lustre. In this session, TotalCAE discusses client case studies using these AWS technologies with TotalCAE to advance the state of the art in CAE simulation.

Summary: The TotalCAE platform empowers engineers to submit hundreds of applications to AWS in just a few clicks. In this demo, learn how TotalCAE uses AWS services including Elastic Fabric Adaptor (EFA), NICE DCV, and Amazon FSx for Lustre and HPC-specific instances like hpc6a.48xlarge. Look at client case studies utilizing these AWS technologies with TotalCAE to advance the state of the art of CAE simulation.

Summary: The Arm-based AWS Graviton2 and AWS Graviton3 processors available on AWS offer high core-counts, high memory-bandwidth, and high floating-point performance. This session shares experiences with benchmarking these processors for HPC-style codes. In addition to lower-level testing of memory bandwidth and latency, learn about multi-node performance for larger applications, including benchmarks from SPEChpc. Also, explore programming these Arm-based processors using a diverse range of the latest parallel programming models and many major open-source and vendor compilers, including standard parallelism in C++ and Fortran.

Summary: In this talk we deep-dive into the origins and technology behind our specialized HPC & machine learning network interface, EFA.

Summary: Join this demo to learn how deploying Enroot and Pyxis for container orchestration on AWS ParallelCluster integrates nicely with Slurm for truly scalable deep learning. Also, discover how GPUDirect RDMA support on Elastic Fabric Adapter (EFA) positively impacts performance.

Summary: Rescale, a leading provider for high-performance computing built for AWS, helps customers utilize AWS services like Amazon S3, Amazon EC2, and NICE DCV for engineering and scientific workloads. In this demo, join Rescale engineer Chris Ramirez to learn how aerospace engineers are accelerating compute-driven innovation with Rescale’s platform, which delivers HPC built for the cloud in collaboration with AWS. Find out how organizations automate full-stack HPC operations in the cloud to optimize the cost performance of computer-aided engineering simulations like CFD and FEA.

Summary: In this talk we deep-dive into the origins and technology behind our specialized HPC & machine learning network interface, EFA.

Summary: Graph Analytics is a form of data analysis that enables us to find and understand complex relationships between data entities. In this presentation, we dive into current work on algorithms and methods that build in network latency tolerance to perform large-scale graph calculations, and discuss results using Elastic Fabric Adapter (EFA), the network interface for Amazon EC2 instances targeted to run applications requiring high levels of inter-node communications on AWS.

Summary: Arm Forge—widely used in HPC labs in traditional environments and the cloud—makes HPC debugging and profiling on Amazon EC2 simple regardless of which processor architecture you use. In this demo, learn how Arm Forge offers you tools that provide seamless access to remote servers, and dive deep on profiling experiences with AWS Graviton3 on HPC applications.

Summary: Arm Forge—widely used in HPC labs in traditional environments and the cloud—makes HPC debugging and profiling on Amazon EC2 simple regardless of which processor architecture you use. In this demo, learn how Arm Forge offers you tools that provide seamless access to remote servers, and dive deep on profiling experiences with AWS Graviton3 on HPC applications.

Summary: This session demonstrates how the NVIDIA HPC SDK activates high performance computing (HPC) on over 200 Amazon EC2 instance types comprised of a variety of Arm or x86 CPUs, with or without GPUs. The NVIDIA HPC SDK is a multi-platform, standards-compliant, vendor-supported solution for HPC application development that supports programming models like OpenMP, OpenACC, and CUDA together with ISO standard parallel features in C++ and Fortran. Its proven compilers, libraries, and software tools support AWS Graviton, x86, POWER, and NVIDIA GPUs to maximize developer productivity, activate hardware platform choice, and facilitate an optimal price-performance solution for HPC applications in the cloud.

Summary: The Arm-based AWS Graviton2 and AWS Graviton3 processors available on AWS offer high core-counts, high memory-bandwidth, and high floating-point performance. This session shares experiences with benchmarking these processors for HPC-style codes. In addition to lower-level testing of memory bandwidth and latency, learn about multi-node performance for larger applications, including benchmarks from SPEChpc. Also, explore programming these Arm-based processors using a diverse range of the latest parallel programming models and many major open-source and vendor compilers, including standard parallelism in C++ and Fortran.

Summary: Arm Forge—widely used in HPC labs in traditional environments and the cloud—makes HPC debugging and profiling on Amazon EC2 simple regardless of which processor architecture you use. In this demo, learn how Arm Forge offers you tools that provide seamless access to remote servers, and dive deep on profiling experiences with AWS Graviton3 on HPC applications.

Summary: Over the last few years, machine learning (ML), more specifically deep learning (DL), has become important for the HPC community. DL uses frameworks like TensorFlow and PyTorch, which use underlying hardware features for better performance. In this session, learn about how DL frameworks use ML-specific Arm architecture features like BFloat16 for faster inference on AWS Graviton3 compared to other platforms.

Summary: Join this demo for an introduction to extending your current on-premises cluster to make use of AWS HPC resources. See a walkthrough on the different requirements for a hybrid architecture, including infrastructure management, security, data transfer, cost, and best practices. Also, discover how to deploy and use a hybrid HPC cluster.

Summary: Run hundreds of thousands of batch computing workloads without managing infrastructure or a job scheduler with AWS Batch. Stop by for live demos including "Introducing AWS Batch Fair Share Scheduling", "Introducing AWS Batch on EKS", "A cloud-native SciOps pipeline for GROMACS", "Run genomics workflows using the Amazon Genomics CLI and AWS Batch"

Summary: Run hundreds of thousands of batch computing workloads without managing infrastructure or a job scheduler with AWS Batch. Stop by for live demos including "Introducing AWS Batch Fair Share Scheduling", "Introducing AWS Batch on EKS", "A cloud-native SciOps pipeline for GROMACS", "Run genomics workflows using the Amazon Genomics CLI and AWS Batch"

Summary: Run hundreds of thousands of batch computing workloads without managing infrastructure or a job scheduler with AWS Batch. Stop by for live demos including "Introducing AWS Batch Fair Share Scheduling", "Introducing AWS Batch on EKS", "A cloud-native SciOps pipeline for GROMACS", "Run genomics workflows using the Amazon Genomics CLI and AWS Batch"

Summary: Whether you are training large machine learning models or running genomic sequencing, EDA, video encoding, big data processing, or autonomous vehicle simulation workloads in the cloud, you’re going to need cloud-native scheduling and orchestration. In this session, learn about AWS Batch, a container-centric, fully managed service for job schedule and AWS compute resource orchestration. Dive deep into AWS Batch major components, such as job definition, job queues, job states, compute environments, instance selection, job throughput, and allocation strategies. Also, discover best practices that can help you build optimized AWS Batch architectures and run AWS Batch at scale.

Summary: Run hundreds of thousands of batch computing workloads without managing infrastructure or a job scheduler with AWS Batch. Stop by for live demos including "Introducing AWS Batch Fair Share Scheduling", "Introducing AWS Batch on EKS", "A cloud-native SciOps pipeline for GROMACS", "Run genomics workflows using the Amazon Genomics CLI and AWS Batch"

Summary: Run hundreds of thousands of batch computing workloads without managing infrastructure or a job scheduler with AWS Batch. Stop by for live demos including "Introducing AWS Batch Fair Share Scheduling", "Introducing AWS Batch on EKS", "A cloud-native SciOps pipeline for GROMACS", "Run genomics workflows using the Amazon Genomics CLI and AWS Batch"

Summary: CryoElectron Microscopy (CryoEM) is revolutionizing the field of structure-based drug design by representing protein structures at that resolutions that are needed for drug discovery. However, CryoEM analysis workflow provides unique challenges for data handling, compute availability, automation, and orchestration. In this demo, learn how Clovertex’s cloud-based scientific discovery platform for structural biologists offers single-click deployment in your VPC so you can scalably and affordably store and process cryoEM data in the cloud. The platform uses 30+ AWS services and brings together the learnings from extensive benchmarking. Empower your scientists to focus on science by reducing cost of analysis and analyzing more structures in a short amount of time.

Summary: Did you know that the Amazon Genomics CLI supports the top 4 worklfow languages for genomics? AGC stands up the AWS infrastructure you need for running workflows written in WDL, Nextflow, CWL, or SnakeMake. In this demo, we will show you how to leverage AGC to run your workflows on top of AWS Batch.

Summary: In this demo, join Clovertex to learn about its scientific research platform and how it can be used to simplify and accelerate structure-based drug design. Also, discover how a cryo-EM pipeline can use the popular CryoSPARC Live application.

Summary: CryoElectron Microscopy (CryoEM) is revolutionizing the field of structure-based drug design by representing protein structures at that resolutions that are needed for drug discovery. However, CryoEM analysis workflow provides unique challenges for data handling, compute availability, automation, and orchestration. In this demo, learn how Clovertex’s cloud-based scientific discovery platform for structural biologists offers single-click deployment in your VPC so you can scalably and affordably store and process cryoEM data in the cloud. The platform uses 30+ AWS services and brings together the learnings from extensive benchmarking. Empower your scientists to focus on science by reducing cost of analysis and analyzing more structures in a short amount of time.

Summary: CryoElectron Microscopy (CryoEM) is revolutionizing the field of structure-based drug design by representing protein structures at that resolutions that are needed for drug discovery. However, CryoEM analysis workflow provides unique challenges for data handling, compute availability, automation, and orchestration. In this demo, learn how Clovertex’s cloud-based scientific discovery platform for structural biologists offers single-click deployment in your VPC so you can scalably and affordably store and process cryoEM data in the cloud. The platform uses 30+ AWS services and brings together the learnings from extensive benchmarking. Empower your scientists to focus on science by reducing cost of analysis and analyzing more structures in a short amount of time.

Summary: Join this session to learn about AWS’s newest storage service: Amazon File Cache, which helps you burst HPC workloads to the cloud. Or, come by for a demo of how to configure Amazon FSx for Lustre performance and capacity for high scale, and native access to S3 data.

Summary: In this live demo, learn about the Parallel Works single-pane-of-glass, multi-cloud environment, which is operated by General Dynamics Information Technology for NOAA’s R&D HPC organization. The presentation demonstrates key aspects of the implementation and optimization of the platform for large-scale HPC-driven R&D. The Parallel Works platform uses 20+ AWS services, including Amazon FSx for Lustre, Amazon S3, Amazon EC2, and Amazon EFS. Dive deep on the end user experience, which incorporates both the familiar on-premises computing environment and the flexibility and innovation of the cloud.

Summary: Join this session to learn about AWS’s newest storage service: Amazon File Cache, which helps you burst HPC workloads to the cloud. Or, come by for a demo of how to configure Amazon FSx for Lustre performance and capacity for high scale, and native access to S3 data.

Summary: The TotalCAE platform empowers engineers to submit hundreds of applications to AWS in just a few clicks. In this demo, learn how TotalCAE uses AWS services including Elastic Fabric Adaptor (EFA), NICE DCV, and Amazon FSx for Lustre and HPC-specific instances like hpc6a.48xlarge. Look at client case studies utilizing these AWS technologies with TotalCAE to advance the state of the art of CAE simulation.

Summary: AWS now has five file systems and a cloud cache. In this session, get an overview of these options, get advice about when to use them, and dive deep into the AWS HPC–specific offerings: Amazon FSx for Lustre, Amazon FSx for OpenZFS, and Amazon File Cache. For each service, discover common use cases and integrations with other key AWS HPC services such as AWS ParallelCluster, AWS Batch, and Elastic Fabric Adapter (EFA).

Summary: Amazon File Cache can accelerate file-based workloads for HPC environments and activate more efficient cloud bursting. Join this demo to learn about this new service and how it’s helping take Lustre mainstream.

Summary: Join this session to learn about AWS’s newest storage service: Amazon File Cache, which helps you burst HPC workloads to the cloud. Or, come by for a demo of how to configure Amazon FSx for Lustre performance and capacity for high scale, and native access to S3 data.

Summary: In this demo, the Atos Nimbix team presents a state-of-the-art container-based solution for multi-site and cloud-based HPC workloads. Learn about this solution, complete with a fully integrated application marketplace featuring curated, ready-to-use click-to-run workflows for simulation applications and AI/ML/DL software.

Summary: In this session, explore how AWS is empowering customers to push the boundaries of computational fluid dynamics (CFD) and helping engineers and designers explore more designs and bring products to market faster. Learn about the move toward GPU-enabled CFD codes, the benefits of cloud elasticity, and the growth of machine learning and deep learning for accelerating CFD simulations.

Summary: In this demo hear from SchedMD, the company behind Slurm workload manager software. Slurm is a free open-source workload manager designed specifically to satisfy the demanding needs of high-performance computing. The software is in widespread use at government laboratories, universities, and companies worldwide and performs workload management for over half of the top 10 systems in the TOP500. Slurm is a scheduler of choice in AWS ParallelCluster.

Summary: Join this demo to learn how deploying Enroot and Pyxis for container orchestration on AWS ParallelCluster integrates nicely with Slurm for truly scalable deep learning. Also, discover how GPUDirect RDMA support on Elastic Fabric Adapter (EFA) positively impacts performance.

Summary: In this demo, the Atos Nimbix team presents a state-of-the-art container-based solution for multi-site and cloud-based HPC workloads. Learn about this solution, complete with a fully integrated application marketplace featuring curated, ready-to-use click-to-run workflows for simulation applications and AI/ML/DL software.

Summary: Over the last few years, machine learning (ML), more specifically deep learning (DL), has become important for the HPC community. DL uses frameworks like TensorFlow and PyTorch, which use underlying hardware features for better performance. In this session, learn about how DL frameworks use ML-specific Arm architecture features like BFloat16 for faster inference on AWS Graviton3 compared to other platforms.

Summary: At ISC 2022, Spack and AWS announced Spack’s first public binary cache, which speeds up scientific software deployment by over 20 times. Come to this session for a detailed look at how developers at LLNL and Kitware used AWS and GitLab CI to achieve continuous deployment of signed, trusted binaries from a high-velocity Spack repository. Learn about the challenges and solutions involved in applying both modern DevOps and AWS computing power to an HPC problem space.

Summary: Launch and manage dynamically scaling high performance computing (HPC) clusters on AWS with AWS ParallelCluster. Stop by for demos including "Introduction to AWS Parallel Cluster", "Introduction to Pcluster Manager", "License Management in AWS ParallelCluster", "AWS ParallelCluster Fast Failover", and "My laptop has 50k cores".

Summary: Graph Analytics is a form of data analysis that enables us to find and understand complex relationships between data entities. In this presentation, we dive into current work on algorithms and methods that build in network latency tolerance to perform large-scale graph calculations, and discuss results using Elastic Fabric Adapter (EFA), the network interface for Amazon EC2 instances targeted to run applications requiring high levels of inter-node communications on AWS.

Summary: Launch and manage dynamically scaling high performance computing (HPC) clusters on AWS with AWS ParallelCluster. Stop by for demos including "Introduction to AWS Parallel Cluster", "Introduction to Pcluster Manager", "License Management in AWS ParallelCluster", "AWS ParallelCluster Fast Failover", and "My laptop has 50k cores".

Summary: In this demo hear from SchedMD, the company behind Slurm workload manager software. Slurm is a free open-source workload manager designed specifically to satisfy the demanding needs of high-performance computing. The software is in widespread use at government laboratories, universities, and companies worldwide and performs workload management for over half of the top 10 systems in the TOP500. Slurm is a scheduler of choice in AWS ParallelCluster.

Summary: In this demo, learn about RONIN, a web application that provides a user-friendly solution to self-service cloud computing via an intuitive interface, making the enormous power of AWS and AWS HPC accessible to even the most novice users. Harnessing the immense scale, power and capability of AWS, RONIN places research applications as well as scalable, on-demand, cutting-edge CPU and GPU processing power directly in the hands of researchers. The RONIN team have worked with leading research organizations and universities to develop a unique software that releases research from traditional on-premises constraints, empowering researchers and scientists to initiate and scale their research cloud computing at the click of a button.

Summary: CryoElectron Microscopy (CryoEM) is revolutionizing the field of structure-based drug design by representing protein structures at that resolutions that are needed for drug discovery. However, CryoEM analysis workflow provides unique challenges for data handling, compute availability, automation, and orchestration. In this demo, learn how Clovertex’s cloud-based scientific discovery platform for structural biologists offers single-click deployment in your VPC so you can scalably and affordably store and process cryoEM data in the cloud. The platform uses 30+ AWS services and brings together the learnings from extensive benchmarking. Empower your scientists to focus on science by reducing cost of analysis and analyzing more structures in a short amount of time.

Summary: In this demo, learn how the QEMIST Cloud platform combines AWS cloud infrastructure; compute and managed services like Amazon EC2, Amazon EKS, and Amazon Aurora; and the recently announced high-performance scaling service Karpenter to build a massively parallel multi-cluster supercomputer for chemistry applications. The computing power of this platform was tested by using it to solve a record-breaking near-exact simulation of Decacene (a molecule with ~200 active electrons in ~600 orbitals) in only 11 hours at merely $2K in cloud costs. This demo shows how cloud computing and the latest developments in AI and quantum computing are shaping the future of high-performance simulations for chemistry, advanced materials, and drug discovery.

Summary: At ISC 2022, Spack and AWS announced Spack’s first public binary cache, which speeds up scientific software deployment by over 20 times. Come to this session for a detailed look at how developers at LLNL and Kitware used AWS and GitLab CI to achieve continuous deployment of signed, trusted binaries from a high-velocity Spack repository. Learn about the challenges and solutions involved in applying both modern DevOps and AWS computing power to an HPC problem space.

Summary: While researchers are increasingly using cloud computing offerings, many aren’t experienced with system administration or cloud service management. This can be challenging for groups without cloud specialists to manage infrastructure and may reduce cloud engagement. Since 2020, the Advanced Research Computing team at the University of British Columbia (UBC) has been using RONIN to deliver a full-service cloud offering to their research community. Come learn about their community, the challenges they faced, and the lessons they learned with the cloud while using the user interfaces on AWS. Hear about their experience with RONIN, what researchers can expect with RONIN, and what’s required to create a workflow in a cloud environment.

Summary: In this talk we deep-dive into the origins and technology behind our specialized HPC & machine learning network interface, EFA.

Summary: In this demo, learn how PCluster Manager and Spack’s binary cache together make a great option for easy design and implementation of highly reproducible environments for scientific research and engineering design.

Summary: In this demo, join Clovertex to learn about its scientific research platform and how it can be used to simplify and accelerate structure-based drug design. Also, discover how a cryo-EM pipeline can use the popular CryoSPARC Live application.

Summary: Organizations in the healthcare and life sciences domain have several applications for on-demand cloud computing. Genome sequencing, clinical trial simulations, and analysis of therapeutic compounds require a scalable, secure, and on-demand computing environment. Service Workbench on AWS is an open-source solution that enables IT teams to create secure portals for researchers on AWS, which combine data management, research tools, and cost visibility without requiring researchers to use the AWS console. In this session, learn about the goals of Service Workbench on AWS and the architecture for HPC customers, see a demo of a Service Workbench HPC prototype with AWS ParallelCluster integration, and hear about the roadmap for 2022 and beyond.

Summary: Launch and manage dynamically scaling high performance computing (HPC) clusters on AWS with AWS ParallelCluster. Stop by for demos including "Introduction to AWS Parallel Cluster", "Introduction to Pcluster Manager", "License Management in AWS ParallelCluster", "AWS ParallelCluster Fast Failover", and "My laptop has 50k cores".

Summary: In this demo, learn about RONIN, a web application that provides a user-friendly solution to self-service cloud computing via an intuitive interface, making the enormous power of AWS and AWS HPC accessible to even the most novice users. Harnessing the immense scale, power and capability of AWS, RONIN places research applications as well as scalable, on-demand, cutting-edge CPU and GPU processing power directly in the hands of researchers. The RONIN team have worked with leading research organizations and universities to develop a unique software that releases research from traditional on-premises constraints, empowering researchers and scientists to initiate and scale their research cloud computing at the click of a button.

Summary: In this demo hear from SchedMD, the company behind Slurm workload manager software. Slurm is a free open-source workload manager designed specifically to satisfy the demanding needs of high-performance computing. The software is in widespread use at government laboratories, universities, and companies worldwide and performs workload management for over half of the top 10 systems in the TOP500. Slurm is a scheduler of choice in AWS ParallelCluster.

Summary: CryoElectron Microscopy (CryoEM) is revolutionizing the field of structure-based drug design by representing protein structures at that resolutions that are needed for drug discovery. However, CryoEM analysis workflow provides unique challenges for data handling, compute availability, automation, and orchestration. In this demo, learn how Clovertex’s cloud-based scientific discovery platform for structural biologists offers single-click deployment in your VPC so you can scalably and affordably store and process cryoEM data in the cloud. The platform uses 30+ AWS services and brings together the learnings from extensive benchmarking. Empower your scientists to focus on science by reducing cost of analysis and analyzing more structures in a short amount of time.

Summary: The Arm-based AWS Graviton2 and AWS Graviton3 processors available on AWS offer high core-counts, high memory-bandwidth, and high floating-point performance. This session shares experiences with benchmarking these processors for HPC-style codes. In addition to lower-level testing of memory bandwidth and latency, learn about multi-node performance for larger applications, including benchmarks from SPEChpc. Also, explore programming these Arm-based processors using a diverse range of the latest parallel programming models and many major open-source and vendor compilers, including standard parallelism in C++ and Fortran.

Summary: In this talk we deep-dive into the origins and technology behind our specialized HPC & machine learning network interface, EFA.

Summary: Join this demo to learn how deploying Enroot and Pyxis for container orchestration on AWS ParallelCluster integrates nicely with Slurm for truly scalable deep learning. Also, discover how GPUDirect RDMA support on Elastic Fabric Adapter (EFA) positively impacts performance.

Summary: Research Gateway provides a Self-Service UI Portal (SaaS and Enterprise models) for Researchers to provision, run analysis and manage costs of complex HPC architecture on AWS with great simplicity. The solution allows users to create an HPC powered MyResearch Cloud in less than 30 min from a standard catalog, with ability to add your own tools. Built on AWS Parallel Cluster with intelligent cost tracking, automation blueprints for workload migrations and project-based data collaboration, researchers can focus on Science and Research Grants funding controls without need for any AWS Console access. Central IT can easily extend the self-service models to a large group of researchers with in-built governance controls and billing for chargebacks.

Summary: Education institutions increasingly use AWS for research, particularly for complex workflows and HTC/HPC workloads. Researchers recognize that the flexibility, breadth of tools, resources, and services offered by AWS provide access to environments that meet their needs on their schedules, but they don’t want to spend time learning about computational infrastructure. This session illustrates how researchers and institutions can accelerate AWS adoption through curated guidance from the AWS research group. Topics include guides and preconfigured AWS tools like AWS ParallelCluster, explainers for software like MATLAB and libraries like Dask that support AWS natively, institutional engagements for on-premises bursting to AWS, research-focused services like Amazon Lightsail and Amazon SageMaker Studio Lab, and proposal development.

Summary: Launch and manage dynamically scaling high performance computing (HPC) clusters on AWS with AWS ParallelCluster. Stop by for demos including "Introduction to AWS Parallel Cluster", "Introduction to Pcluster Manager", "License Management in AWS ParallelCluster", "AWS ParallelCluster Fast Failover", and "My laptop has 50k cores".

Summary: In this demo, learn about RONIN, a web application that provides a user-friendly solution to self-service cloud computing via an intuitive interface, making the enormous power of AWS and AWS HPC accessible to even the most novice users. Harnessing the immense scale, power and capability of AWS, RONIN places research applications as well as scalable, on-demand, cutting-edge CPU and GPU processing power directly in the hands of researchers. The RONIN team have worked with leading research organizations and universities to develop a unique software that releases research from traditional on-premises constraints, empowering researchers and scientists to initiate and scale their research cloud computing at the click of a button.

Summary: CryoElectron Microscopy (CryoEM) is revolutionizing the field of structure-based drug design by representing protein structures at that resolutions that are needed for drug discovery. However, CryoEM analysis workflow provides unique challenges for data handling, compute availability, automation, and orchestration. In this demo, learn how Clovertex’s cloud-based scientific discovery platform for structural biologists offers single-click deployment in your VPC so you can scalably and affordably store and process cryoEM data in the cloud. The platform uses 30+ AWS services and brings together the learnings from extensive benchmarking. Empower your scientists to focus on science by reducing cost of analysis and analyzing more structures in a short amount of time.

Summary: Over the last few years, machine learning (ML), more specifically deep learning (DL), has become important for the HPC community. DL uses frameworks like TensorFlow and PyTorch, which use underlying hardware features for better performance. In this session, learn about how DL frameworks use ML-specific Arm architecture features like BFloat16 for faster inference on AWS Graviton3 compared to other platforms.

Summary: In this talk we deep-dive into the origins and technology behind our specialized HPC & machine learning network interface, EFA.

Summary: In this demo, learn how DXC put together a managed service on AWS to utilize all the benefits of HPC in the cloud with a focus on CAA-II tooling. The presentation covers the components of a hybrid cloud as well as the challenges of legacy and on prem and heterogeneity, and control of the cloud—how do you leverage the power of the cloud while preserving budgetary, security, and export control? Also learn what makes a great service and find out about automation, monitoring integration, and tooling.

Summary: In this demo, learn about RONIN, a web application that provides a user-friendly solution to self-service cloud computing via an intuitive interface, making the enormous power of AWS and AWS HPC accessible to even the most novice users. Harnessing the immense scale, power and capability of AWS, RONIN places research applications as well as scalable, on-demand, cutting-edge CPU and GPU processing power directly in the hands of researchers. The RONIN team have worked with leading research organizations and universities to develop a unique software that releases research from traditional on-premises constraints, empowering researchers and scientists to initiate and scale their research cloud computing at the click of a button.

Summary: In 2016, Alces Flight launched one of the first end user marketplace products specifically for HPC. In this demo, learn about the latest updates to the full-service cloud HPC management tool Alces Flight Center and see how Alces has leveled up all aspects of cloud engagement—from users, to environments, to cost control. And, making its debut at SC22, see the successor to Alces’s cloud HPC product built entirely for end users, now ready to take on today’s research challenges. The demo also features real-life cloud-only and hybrid HPC work case studies that were recognized by HPCWire’s Readers’ Choice in 2020 and 2022.

Summary: Rescale, a leading provider for high-performance computing built for AWS, helps customers utilize AWS services like Amazon S3, Amazon EC2, and NICE DCV for engineering and scientific workloads. In this demo, join Rescale engineer Chris Ramirez to learn how aerospace engineers are accelerating compute-driven innovation with Rescale’s platform, which delivers HPC built for the cloud in collaboration with AWS. Find out how organizations automate full-stack HPC operations in the cloud to optimize the cost performance of computer-aided engineering simulations like CFD and FEA.

Summary: Research Gateway provides a self-service UI portal (featuring SaaS and enterprise models) for researchers to provision, run analysis, and manage the costs of complex HPC architecture on AWS. The solution allows users to create an HPC-powered MyResearch Cloud in less than 30 minutes from a standard catalog, with the ability to incorporate their own tools. Built on AWS ParallelCluster, Research Gateway includes intelligent cost tracking, blueprints for automated migration, and project-based data collaboration and allows researchers to focus on science and research grants without the need for AWS Console access. Central IT can easily extend the self-service models to a group of researchers with built-in governance controls and chargebacks billing.

Summary: In this live demo, learn about the Parallel Works single-pane-of-glass, multi-cloud environment, which is operated by General Dynamics Information Technology for NOAA’s R&D HPC organization. The presentation demonstrates key aspects of the implementation and optimization of the platform for large-scale HPC-driven R&D. The Parallel Works platform uses 20+ AWS services, including Amazon FSx for Lustre, Amazon S3, Amazon EC2, and Amazon EFS. Dive deep on the end user experience, which incorporates both the familiar on-premises computing environment and the flexibility and innovation of the cloud.

Summary: While researchers are increasingly using cloud computing offerings, many aren’t experienced with system administration or cloud service management. This can be challenging for groups without cloud specialists to manage infrastructure and may reduce cloud engagement. Since 2020, the Advanced Research Computing team at the University of British Columbia (UBC) has been using RONIN to deliver a full-service cloud offering to their research community. Come learn about their community, the challenges they faced, and the lessons they learned with the cloud while using the user interfaces on AWS. Hear about their experience with RONIN, what researchers can expect with RONIN, and what’s required to create a workflow in a cloud environment.

Summary: In this session, learn about the benefits of running Altair Radioss on Amazon EC2 Hpc6a instances powered by third-generation AMD EPYC processors. For numerous, diverse industries worldwide, Radioss is a versatile and comprehensive high-performance explicit finite element (FE) solver, based on solid and fluid dynamics. It is a standard, go-to tool to accurately predict dynamic, transient-loading effects on structures and products to improve safety and survivability and to design more robust products. Radioss on Hpc6a instances delivers impressive performance on premises. Discover the scaling and performance improvements achieved from Radioss running on Hpc6a instances.

Summary: Organizations in the healthcare and life sciences domain have several applications for on-demand cloud computing. Genome sequencing, clinical trial simulations, and analysis of therapeutic compounds require a scalable, secure, and on-demand computing environment. Service Workbench on AWS is an open-source solution that enables IT teams to create secure portals for researchers on AWS, which combine data management, research tools, and cost visibility without requiring researchers to use the AWS console. In this session, learn about the goals of Service Workbench on AWS and the architecture for HPC customers, see a demo of a Service Workbench HPC prototype with AWS ParallelCluster integration, and hear about the roadmap for 2022 and beyond.

Summary: In this demo, learn about a new service powered by AWS that Ansys is preparing to launch: Ansys Gateway. Ansys Gateway enables simpler access and deployment for several industry-leading Ansys simulations on AWS. Ansys is also performance-testing several of its products on Amazon EC2 instances.

Summary: In this live demo, learn about the Parallel Works single-pane-of-glass, multi-cloud environment, which is operated by General Dynamics Information Technology for NOAA’s R&D HPC organization. The presentation demonstrates key aspects of the implementation and optimization of the platform for large-scale HPC-driven R&D. The Parallel Works platform uses 20+ AWS services, including Amazon FSx for Lustre, Amazon S3, Amazon EC2, and Amazon EFS. Dive deep on the end user experience, which incorporates both the familiar on-premises computing environment and the flexibility and innovation of the cloud.

Summary: The classification of workloads into precise execution patterns is difficult and can lead to suboptimal execution in order to avoid the proliferation of applications, platforms, and techniques. What if you could choose a unified and intelligent workload submission platform to help you take advtange of the speed and scalability of the AWS Cloud? Regardless of your workload type, if it is compute-intensive, YellowDog on AWS can help you manage it. YellowDog is a fast, scalable, simple, Spot-friendly extensible scheduler and provisioning platform for hybrid and cloud. Concentrate on the insights and outcomes your workloads generate, and let YellowDog on AWS take care of everything else.

Summary: What are the best instances I can use to run my HPC workload? What is the best instance balance? What is the optimal elasticity strategy for my workload? What is the best use of my budget? How can I limit waste? Based on OKA, a leading data science platform for hybrid HPC, CloudSHaper provides answers to these questions and a lot more. During this demo, the CloudSHaper team provides a guided tour of the platform.

Summary: In 2016, Alces Flight launched one of the first end user marketplace products specifically for HPC. In this demo, learn about the latest updates to the full-service cloud HPC management tool Alces Flight Center and see how Alces has leveled up all aspects of cloud engagement—from users, to environments, to cost control. And, making its debut at SC22, see the successor to Alces’s cloud HPC product built entirely for end users, now ready to take on today’s research challenges. The demo also features real-life cloud-only and hybrid HPC work case studies that were recognized by HPCWire’s Readers’ Choice in 2020 and 2022.

Summary: In this demo, learn about RONIN, a web application that provides a user-friendly solution to self-service cloud computing via an intuitive interface, making the enormous power of AWS and AWS HPC accessible to even the most novice users. Harnessing the immense scale, power and capability of AWS, RONIN places research applications as well as scalable, on-demand, cutting-edge CPU and GPU processing power directly in the hands of researchers. The RONIN team have worked with leading research organizations and universities to develop a unique software that releases research from traditional on-premises constraints, empowering researchers and scientists to initiate and scale their research cloud computing at the click of a button.

Summary: Rescale, a leading provider for high-performance computing built for AWS, helps customers utilize AWS services like Amazon S3, Amazon EC2, and NICE DCV for engineering and scientific workloads. In this demo, join Rescale engineer Chris Ramirez to learn how aerospace engineers are accelerating compute-driven innovation with Rescale’s platform, which delivers HPC built for the cloud in collaboration with AWS. Find out how organizations automate full-stack HPC operations in the cloud to optimize the cost performance of computer-aided engineering simulations like CFD and FEA.

Summary: In this demo, learn about a new service powered by AWS that Ansys is preparing to launch: Ansys Gateway. Ansys Gateway enables simpler access and deployment for several industry-leading Ansys simulations on AWS. Ansys is also performance-testing several of its products on Amazon EC2 instances.

Summary: Rescale is a premier provider for high performance computing built for AWS. Organizations across all industries have found immense value in using the AWS full-service platform. Join this session to learn how to quickly and efficiently run your various simulation packages on infrastructure like C6i instances, powered by Intel’s Xeon Ice Lake processor.

Summary: Research Gateway provides a Self-Service UI Portal (SaaS and Enterprise models) for Researchers to provision, run analysis and manage costs of complex HPC architecture on AWS with great simplicity. The solution allows users to create an HPC powered MyResearch Cloud in less than 30 min from a standard catalog, with ability to add your own tools. Built on AWS Parallel Cluster with intelligent cost tracking, automation blueprints for workload migrations and project-based data collaboration, researchers can focus on Science and Research Grants funding controls without need for any AWS Console access. Central IT can easily extend the self-service models to a large group of researchers with in-built governance controls and billing for chargebacks.

Summary: Research Gateway provides a self-service UI portal (featuring SaaS and enterprise models) for researchers to provision, run analysis, and manage the costs of complex HPC architecture on AWS. The solution allows users to create an HPC-powered MyResearch Cloud in less than 30 minutes from a standard catalog, with the ability to incorporate their own tools. Built on AWS ParallelCluster, Research Gateway includes intelligent cost tracking, blueprints for automated migration, and project-based data collaboration and allows researchers to focus on science and research grants without the need for AWS Console access. Central IT can easily extend the self-service models to a group of researchers with built-in governance controls and chargebacks billing.

Summary: In this demo, learn about RONIN, a web application that provides a user-friendly solution to self-service cloud computing via an intuitive interface, making the enormous power of AWS and AWS HPC accessible to even the most novice users. Harnessing the immense scale, power and capability of AWS, RONIN places research applications as well as scalable, on-demand, cutting-edge CPU and GPU processing power directly in the hands of researchers. The RONIN team have worked with leading research organizations and universities to develop a unique software that releases research from traditional on-premises constraints, empowering researchers and scientists to initiate and scale their research cloud computing at the click of a button.

Summary: Rescale, a leading provider for high-performance computing built for AWS, helps customers utilize AWS services like Amazon S3, Amazon EC2, and NICE DCV for engineering and scientific workloads. In this demo, join Rescale engineer Chris Ramirez to learn how aerospace engineers are accelerating compute-driven innovation with Rescale’s platform, which delivers HPC built for the cloud in collaboration with AWS. Find out how organizations automate full-stack HPC operations in the cloud to optimize the cost performance of computer-aided engineering simulations like CFD and FEA.

Summary: Come learn about the AWS Center for Quantum Computing, Amazon Quantum Solutions Lab, AWS Partner Network, and Amazon Braket.

Summary: Covalent is an open-source Pythonic tool for heterogeneous workflows (HPC and quantum), designed specifically to make computational research more scalable, manageable, and reproducible. Covalent provides a rich and intuitive browser-based interface that empowers users to visualize and interact with their workflows. Covalent integrates with a number of AWS cloud services that allow users to dispatch their workflows for remote execution on diverse compute backends. Using Covalent AWS plugins, users can prototype, iterate and rapidly optimize their workflows. Users can also self-host Covalent within their own AWS infrastructure. This demo highlights several use cases where Covalent can be used in conjunction with AWS compute resources.

Summary: In this demo, learn how the QEMIST Cloud platform combines AWS cloud infrastructure; compute and managed services like Amazon EC2, Amazon EKS, and Amazon Aurora; and the recently announced high-performance scaling service Karpenter to build a massively parallel multi-cluster supercomputer for chemistry applications. The computing power of this platform was tested by using it to solve a record-breaking near-exact simulation of Decacene (a molecule with ~200 active electrons in ~600 orbitals) in only 11 hours at merely $2K in cloud costs. This demo shows how cloud computing and the latest developments in AI and quantum computing are shaping the future of high-performance simulations for chemistry, advanced materials, and drug discovery.

Summary: IonQ's quantum computer is the only trapped ion system available on Amazon Braket and delivers unparalleled performance for today's quantum algorithms. Matt Keesan, VP of Product Development will be discussing our technology and opportunities for near term applications of this transformative computational approach. Additionally, Matt will be discussing how to access IonQ's quantum computer through the Amazon Braket service.

Summary: Multiverse Computing—a leading deep-tech company developing quantum software—has developed Singularity Optimization, a Python library used for modeling and solving generic optimization problems with quantum and quantum-inspired solvers. Optimization is one of the most promising use cases for quantum computing, which offers a way to move beyond the limitations of conventional computers. Multiverse and commercial quantum hardware providers demonstrate already today that real-life industrial optimization problems can be solved efficiently by quantum means. In this demo, learn about numerous examples of Singularity Optimization at work on AWS, and dive into a straightforward interface that requires no prior quantum experience or knowledge.

Summary: This hands-on demo shows visitors how to use the qBraid platform and how they can run their first quantum circuits in real time. More experienced users can learn how to run quantum jobs built using Qiskit (IBM) on Amazon Braket. Customers experienced in Qiskit can explore Amazon Braket and many quantum computers available on the AWS platform.

Summary: In this fast-changing world, staying ahead of the competition is imperative. Quantum Computing Inc. (QCi), working with Amazon Braket, shows you how quantum computing is available today for many use cases. Quantum computing has been limited to small problems, but the recent announcement from QCi about Entropy Quantum Computing changes the HPC landscape. Attend this demo to learn how Entropy Quantum Computing can impact your organization.

Summary: In this demo, QuEra introduces tools and services for quantum computing. Learn about neutral atom–based analog quantum processors and how they are operated. Explore hardware-efficient application directions and how they can be efficiently emulated using AWS HPC resources via QuEra's package, Bloqade.

Summary: In this demo, learn how to eliminate the complexity of accessing, applying, and managing the flood of emerging compute technologies by bringing them together into one platform. Strangeworks offers a hardware-agnostic and software-inclusive platform that acts as a central hub for advanced computing hardware and software vendors and is built specifically for enterprises, academia, researchers, scientists, and engineers to help them solve their most complex research and development challenges. Learn how to use the Strangeworks service library, manipulate quantum code, and generate real results using third-party frameworks and languages. Explore real-world examples from customers and partners, powered by the Strangeworks platform.

Summary: It may seem like science fiction, but quantum computing is here—and it’s available on the AWS Cloud. Although production workloads aren’t available yet, Amazon Braket gives you the ability to get quantum-ready by exploring the capabilities of a range of current quantum hardware, including superconducting qubits, ion traps, photonic qubits, and Rydberg atom arrays, with more being added regularly. Join this session for an introduction to quantum computing, learn about how the Braket service works, and hear AWS Partners discuss the novel solutions they’ve built on Braket.

Summary: Come learn about the AWS Center for Quantum Computing, Amazon Quantum Solutions Lab, AWS Partner Network, and Amazon Braket.

Summary: In this demo, learn how to eliminate the complexity of accessing, applying, and managing the flood of emerging compute technologies by bringing them together into one platform. Strangeworks offers a hardware-agnostic and software-inclusive platform that acts as a central hub for advanced computing hardware and software vendors and is built specifically for enterprises, academia, researchers, scientists, and engineers to help them solve their most complex research and development challenges. Learn how to use the Strangeworks service library, manipulate quantum code, and generate real results using third-party frameworks and languages. Explore real-world examples from customers and partners, powered by the Strangeworks platform.

Summary: In this demo, QuEra introduces tools and services for quantum computing. Learn about neutral atom–based analog quantum processors and how they are operated. Explore hardware-efficient application directions and how they can be efficiently emulated using AWS HPC resources via QuEra's package, Bloqade.

Summary: In this fast-changing world, staying ahead of the competition is imperative. Quantum Computing Inc. (QCi), working with Amazon Braket, shows you how quantum computing is available today for many use cases. Quantum computing has been limited to small problems, but the recent announcement from QCi about Entropy Quantum Computing changes the HPC landscape. Attend this demo to learn how Entropy Quantum Computing can impact your organization.

Summary: This hands-on demo shows visitors how to use the qBraid platform and how they can run their first quantum circuits in real time. More experienced users can learn how to run quantum jobs built using Qiskit (IBM) on Amazon Braket. Customers experienced in Qiskit can explore Amazon Braket and many quantum computers available on the AWS platform.

Summary: Multiverse Computing—a leading deep-tech company developing quantum software—has developed Singularity Optimization, a Python library used for modeling and solving generic optimization problems with quantum and quantum-inspired solvers. Optimization is one of the most promising use cases for quantum computing, which offers a way to move beyond the limitations of conventional computers. Multiverse and commercial quantum hardware providers demonstrate already today that real-life industrial optimization problems can be solved efficiently by quantum means. In this demo, learn about numerous examples of Singularity Optimization at work on AWS, and dive into a straightforward interface that requires no prior quantum experience or knowledge.

Summary: IonQ's quantum computer is the only trapped ion system available on Amazon Braket and delivers unparalleled performance for today's quantum algorithms. Matt Keesan, VP of Product Development will be discussing our technology and opportunities for near term applications of this transformative computational approach. Additionally, Matt will be discussing how to access IonQ's quantum computer through the Amazon Braket service.

Summary: In this demo, learn how the QEMIST Cloud platform combines AWS cloud infrastructure; compute and managed services like Amazon EC2, Amazon EKS, and Amazon Aurora; and the recently announced high-performance scaling service Karpenter to build a massively parallel multi-cluster supercomputer for chemistry applications. The computing power of this platform was tested by using it to solve a record-breaking near-exact simulation of Decacene (a molecule with ~200 active electrons in ~600 orbitals) in only 11 hours at merely $2K in cloud costs. This demo shows how cloud computing and the latest developments in AI and quantum computing are shaping the future of high-performance simulations for chemistry, advanced materials, and drug discovery.

Summary: Covalent is an open-source Pythonic tool for heterogeneous workflows (HPC and quantum), designed specifically to make computational research more scalable, manageable, and reproducible. Covalent provides a rich and intuitive browser-based interface that empowers users to visualize and interact with their workflows. Covalent integrates with a number of AWS cloud services that allow users to dispatch their workflows for remote execution on diverse compute backends. Using Covalent AWS plugins, users can prototype, iterate and rapidly optimize their workflows. Users can also self-host Covalent within their own AWS infrastructure. This demo highlights several use cases where Covalent can be used in conjunction with AWS compute resources.

Summary: Come learn about the AWS Center for Quantum Computing, Amazon Quantum Solutions Lab, AWS Partner Network, and Amazon Braket.

Summary: It's relatively easy to explain how popular quantum algorithms work - how to perform a Grover search, how to optimize an asset portfolio, how to analyze credit risks, and even how Shor's algorithm works. It is much more difficult, however, to implement these algorithms in a generalized way, to allow them to run at scale, and code them in an efficient and hardware-aware way. Come see our demo and experience a radically better way to develop quantum software. Our platform automatically creates optimized and hardware-aware circuits from high-level functional models. It automates and simplifies the difficult process of creating quantum algorithms allowing you to create 100+ qubit software in minutes.

Summary: In this demo, learn how the QEMIST Cloud platform combines AWS cloud infrastructure; compute and managed services like Amazon EC2, Amazon EKS, and Amazon Aurora; and the recently announced high-performance scaling service Karpenter to build a massively parallel multi-cluster supercomputer for chemistry applications. The computing power of this platform was tested by using it to solve a record-breaking near-exact simulation of Decacene (a molecule with ~200 active electrons in ~600 orbitals) in only 11 hours at merely $2K in cloud costs. This demo shows how cloud computing and the latest developments in AI and quantum computing are shaping the future of high-performance simulations for chemistry, advanced materials, and drug discovery.

Summary: This hands-on demo shows visitors how to use the qBraid platform and how they can run their first quantum circuits in real time. More experienced users can learn how to run quantum jobs built using Qiskit (IBM) on Amazon Braket. Customers experienced in Qiskit can explore Amazon Braket and many quantum computers available on the AWS platform.

Summary: In this demo, QuEra introduces tools and services for quantum computing. Learn about neutral atom–based analog quantum processors and how they are operated. Explore hardware-efficient application directions and how they can be efficiently emulated using AWS HPC resources via QuEra's package, Bloqade.

Summary: It may seem like science fiction, but quantum computing is here—and it’s available on the AWS Cloud. Although production workloads aren’t available yet, Amazon Braket gives you the ability to get quantum-ready by exploring the capabilities of a range of current quantum hardware, including superconducting qubits, ion traps, photonic qubits, and Rydberg atom arrays, with more being added regularly. Join this session for an introduction to quantum computing, learn about how the Braket service works, and hear AWS Partners discuss the novel solutions they’ve built on Braket.

Summary: In this demo, learn about RONIN, a web application that provides a user-friendly solution to self-service cloud computing via an intuitive interface, making the enormous power of AWS and AWS HPC accessible to even the most novice users. Harnessing the immense scale, power and capability of AWS, RONIN places research applications as well as scalable, on-demand, cutting-edge CPU and GPU processing power directly in the hands of researchers. The RONIN team have worked with leading research organizations and universities to develop a unique software that releases research from traditional on-premises constraints, empowering researchers and scientists to initiate and scale their research cloud computing at the click of a button.

Summary: Research Gateway provides a self-service UI portal (featuring SaaS and enterprise models) for researchers to provision, run analysis, and manage the costs of complex HPC architecture on AWS. The solution allows users to create an HPC-powered MyResearch Cloud in less than 30 minutes from a standard catalog, with the ability to incorporate their own tools. Built on AWS ParallelCluster, Research Gateway includes intelligent cost tracking, blueprints for automated migration, and project-based data collaboration and allows researchers to focus on science and research grants without the need for AWS Console access. Central IT can easily extend the self-service models to a group of researchers with built-in governance controls and chargebacks billing.

Summary: While researchers are increasingly using cloud computing offerings, many aren’t experienced with system administration or cloud service management. This can be challenging for groups without cloud specialists to manage infrastructure and may reduce cloud engagement. Since 2020, the Advanced Research Computing team at the University of British Columbia (UBC) has been using RONIN to deliver a full-service cloud offering to their research community. Come learn about their community, the challenges they faced, and the lessons they learned with the cloud while using the user interfaces on AWS. Hear about their experience with RONIN, what researchers can expect with RONIN, and what’s required to create a workflow in a cloud environment.

Summary: Organizations in the healthcare and life sciences domain have several applications for on-demand cloud computing. Genome sequencing, clinical trial simulations, and analysis of therapeutic compounds require a scalable, secure, and on-demand computing environment. Service Workbench on AWS is an open-source solution that enables IT teams to create secure portals for researchers on AWS, which combine data management, research tools, and cost visibility without requiring researchers to use the AWS console. In this session, learn about the goals of Service Workbench on AWS and the architecture for HPC customers, see a demo of a Service Workbench HPC prototype with AWS ParallelCluster integration, and hear about the roadmap for 2022 and beyond.

Summary: In this demo, learn about RONIN, a web application that provides a user-friendly solution to self-service cloud computing via an intuitive interface, making the enormous power of AWS and AWS HPC accessible to even the most novice users. Harnessing the immense scale, power and capability of AWS, RONIN places research applications as well as scalable, on-demand, cutting-edge CPU and GPU processing power directly in the hands of researchers. The RONIN team have worked with leading research organizations and universities to develop a unique software that releases research from traditional on-premises constraints, empowering researchers and scientists to initiate and scale their research cloud computing at the click of a button.

Summary: Ray tracing for scientific data provides efficient and realistic lighting and materials for enhanced visual analysis. Recent advances in hardware and software now provide interactive ray tracing on both CPU and GPU platforms. To fully exploit interactive ray tracing in remote computing contexts, researchers require a performant framework that provides an interactive experience even in high latency and low bandwidth contexts. This chalk talk presents how the Texas Advanced Computing Center (TACC) uses NICE DCV and the Intel oneAPI Rendering Toolkit to provide interactive visual analysis.

Summary: Research Gateway provides a Self-Service UI Portal (SaaS and Enterprise models) for Researchers to provision, run analysis and manage costs of complex HPC architecture on AWS with great simplicity. The solution allows users to create an HPC powered MyResearch Cloud in less than 30 min from a standard catalog, with ability to add your own tools. Built on AWS Parallel Cluster with intelligent cost tracking, automation blueprints for workload migrations and project-based data collaboration, researchers can focus on Science and Research Grants funding controls without need for any AWS Console access. Central IT can easily extend the self-service models to a large group of researchers with in-built governance controls and billing for chargebacks.

Summary: Education institutions increasingly use AWS for research, particularly for complex workflows and HTC/HPC workloads. Researchers recognize that the flexibility, breadth of tools, resources, and services offered by AWS provide access to environments that meet their needs on their schedules, but they don’t want to spend time learning about computational infrastructure. This session illustrates how researchers and institutions can accelerate AWS adoption through curated guidance from the AWS research group. Topics include guides and preconfigured AWS tools like AWS ParallelCluster, explainers for software like MATLAB and libraries like Dask that support AWS natively, institutional engagements for on-premises bursting to AWS, research-focused services like Amazon Lightsail and Amazon SageMaker Studio Lab, and proposal development.

Summary: The Global Data Egress Waiver (GDEW) program has been around at AWS for many years. In this session, learn how the program works and how it reduces the risk that academic and government research organizations will ever see outbound data charges on their bills.

Summary: Research Gateway provides a self-service UI portal (featuring SaaS and enterprise models) for researchers to provision, run analysis, and manage the costs of complex HPC architecture on AWS. The solution allows users to create an HPC-powered MyResearch Cloud in less than 30 minutes from a standard catalog, with the ability to incorporate their own tools. Built on AWS ParallelCluster, Research Gateway includes intelligent cost tracking, blueprints for automated migration, and project-based data collaboration and allows researchers to focus on science and research grants without the need for AWS Console access. Central IT can easily extend the self-service models to a group of researchers with built-in governance controls and chargebacks billing.

Summary: In this demo, learn about RONIN, a web application that provides a user-friendly solution to self-service cloud computing via an intuitive interface, making the enormous power of AWS and AWS HPC accessible to even the most novice users. Harnessing the immense scale, power and capability of AWS, RONIN places research applications as well as scalable, on-demand, cutting-edge CPU and GPU processing power directly in the hands of researchers. The RONIN team have worked with leading research organizations and universities to develop a unique software that releases research from traditional on-premises constraints, empowering researchers and scientists to initiate and scale their research cloud computing at the click of a button.

Summary: In this demo hear from SchedMD, the company behind Slurm workload manager software. Slurm is a free open-source workload manager designed specifically to satisfy the demanding needs of high-performance computing. The software is in widespread use at government laboratories, universities, and companies worldwide and performs workload management for over half of the top 10 systems in the TOP500. Slurm is a scheduler of choice in AWS ParallelCluster.

Summary: The classification of workloads into precise execution patterns is difficult and can lead to suboptimal execution in order to avoid the proliferation of applications, platforms, and techniques. What if you could choose a unified and intelligent workload submission platform to help you take advtange of the speed and scalability of the AWS Cloud? Regardless of your workload type, if it is compute-intensive, YellowDog on AWS can help you manage it. YellowDog is a fast, scalable, simple, Spot-friendly extensible scheduler and provisioning platform for hybrid and cloud. Concentrate on the insights and outcomes your workloads generate, and let YellowDog on AWS take care of everything else.

Summary: Workload schedulers are a necessary component of a HPC cluster. In this demo, learn about IBM Spectrum LSF, a popular choice for HPC workload. Over the past 30 years, LSF has evolved from being just a workload scheduler to represent an entire suite of capabilities covering the lifecycle of HPC jobs. Some customers have run on-premises HPC workloads with LSF for decades and wish to have a seamless experience as they adopt cloud while retaining legacy tools, flows and methodologies, and most importantly, time and resources. IBM LSF has delivered for customers desiring ease in license purchase process, HPC cluster deployment, and operations.

Summary: In this demo hear from SchedMD, the company behind Slurm workload manager software. Slurm is a free open-source workload manager designed specifically to satisfy the demanding needs of high-performance computing. The software is in widespread use at government laboratories, universities, and companies worldwide and performs workload management for over half of the top 10 systems in the TOP500. Slurm is a scheduler of choice in AWS ParallelCluster.

Summary: Join this demo to learn how deploying Enroot and Pyxis for container orchestration on AWS ParallelCluster integrates nicely with Slurm for truly scalable deep learning. Also, discover how GPUDirect RDMA support on Elastic Fabric Adapter (EFA) positively impacts performance.

Summary: Workload schedulers are a necessary component of a HPC cluster. In this demo, learn about IBM Spectrum LSF, a popular choice for HPC workload. Over the past 30 years, LSF has evolved from being just a workload scheduler to represent an entire suite of capabilities covering the lifecycle of HPC jobs. Some customers have run on-premises HPC workloads with LSF for decades and wish to have a seamless experience as they adopt cloud while retaining legacy tools, flows and methodologies, and most importantly, time and resources. IBM LSF has delivered for customers desiring ease in license purchase process, HPC cluster deployment, and operations.

Summary: Join this session to learn about AWS’s newest storage service: Amazon File Cache, which helps you burst HPC workloads to the cloud. Or, come by for a demo of how to configure Amazon FSx for Lustre performance and capacity for high scale, and native access to S3 data.

Summary: In this live demo, learn about the Parallel Works single-pane-of-glass, multi-cloud environment, which is operated by General Dynamics Information Technology for NOAA’s R&D HPC organization. The presentation demonstrates key aspects of the implementation and optimization of the platform for large-scale HPC-driven R&D. The Parallel Works platform uses 20+ AWS services, including Amazon FSx for Lustre, Amazon S3, Amazon EC2, and Amazon EFS. Dive deep on the end user experience, which incorporates both the familiar on-premises computing environment and the flexibility and innovation of the cloud.

Summary: Join this session to learn about AWS’s newest storage service: Amazon File Cache, which helps you burst HPC workloads to the cloud. Or, come by for a demo of how to configure Amazon FSx for Lustre performance and capacity for high scale, and native access to S3 data.

Summary: Engineers are using HPC simulations to accelerate the time needed to design and validate new products. To achieve this goal, they need not only to run HPC simulations faster but also to easily modify the input files and visualize the results using their preferred CAD and post-processing software that usually runs on the Windows OS. This demo covers a Windows VDI solution that can create the infrastructure for your users to authenticate and connect to virtual Windows sessions with Amazon AppStream 2.0, submit their HPC jobs to a Linux cluster, and view the output files as they are being written out on an Amazon FSx for NetApp ONTAP file system.

Summary: The TotalCAE platform empowers engineers to submit hundreds of applications to AWS in just a few clicks. In this demo, learn how TotalCAE uses AWS services including Elastic Fabric Adaptor (EFA), NICE DCV, and Amazon FSx for Lustre and HPC-specific instances like hpc6a.48xlarge. Look at client case studies utilizing these AWS technologies with TotalCAE to advance the state of the art of CAE simulation.

Summary: AWS now has five file systems and a cloud cache. In this session, get an overview of these options, get advice about when to use them, and dive deep into the AWS HPC–specific offerings: Amazon FSx for Lustre, Amazon FSx for OpenZFS, and Amazon File Cache. For each service, discover common use cases and integrations with other key AWS HPC services such as AWS ParallelCluster, AWS Batch, and Elastic Fabric Adapter (EFA).

Summary: Amazon File Cache can accelerate file-based workloads for HPC environments and activate more efficient cloud bursting. Join this demo to learn about this new service and how it’s helping take Lustre mainstream.

Summary: Join this session to learn about AWS’s newest storage service: Amazon File Cache, which helps you burst HPC workloads to the cloud. Or, come by for a demo of how to configure Amazon FSx for Lustre performance and capacity for high scale, and native access to S3 data.

Summary: Engineers are using HPC simulations to accelerate the time needed to design and validate new products. To achieve this goal, they need not only to run HPC simulations faster but also to easily modify the input files and visualize the results using their preferred CAD and post-processing software that usually runs on the Windows OS. This demo covers a Windows VDI solution that can create the infrastructure for your users to authenticate and connect to virtual Windows sessions with Amazon AppStream 2.0, submit their HPC jobs to a Linux cluster, and view the output files as they are being written out on an Amazon FSx for NetApp ONTAP file system.

Summary: Visit our NICE DCV demo station for live demonstrations including "NICE DCV in HPC on AWS", "High-resolution showcase with NICE DCV" and "Streaming High-Performance Cloud Gaming with DCV".

Summary: When providing access to VDI, it is a best practice to have the remote machines not be public facing. Providing secure access to DCV servers that are not public facing can be achieved with Connection Gateway. In this presentation, we will talk about how Connection Gateway can be used to lock down DCV environments.

Summary: Visit our NICE DCV demo station for live demonstrations including "NICE DCV in HPC on AWS", "High-resolution showcase with NICE DCV" and "Streaming High-Performance Cloud Gaming with DCV".