Delivering ASIC-like performance, Speedster7t™ FPGAs are highly configurable, highly flexible, data compute engines. Built with high-performance SerDes, high-bandwidth memory interfaces, and high-speed PCIe Gen5 ports, Speedster7t FPGAs feature a massive parallel array of machine learning processors (MLP) blocks to give the highest performance/watt and lowest cost solution for HPC applications.
In the data center and in the cloud, Altair is the industry leader in high-performance, high-throughput computing solutions that help you orchestrate, visualize, optimize, and analyze your most demanding workloads. Top500 systems and small to mid-sized computing environments alike rely on Altair to accelerate innovation and keep infrastructure running smoothly.
The Atos Quantum Learning Machine (Atos QLM) is a complete on premise programming and simulation environment that can be used as a universal front-end for quantum processors. It is dedicated to the development of quantum software, training and experimentation.
The Atos BullSequana X series of servers provide maximum performance, flexibility, scalability and efficiency! They cover the widest spectrum of applications and offer an infinitely adaptable response with numerous combinations. The Atos BullSequana X series empower you to take the full advantage of supercomputing regardless of your needs!
Our latest accelerator cards and modules for high-performance compute, network and storage include the latest FPGAs from Achronix, Intel and Xilinx. These products include the 7nm Achronix Speedster7t FPGA, Intel Agilex, and a NVMe U.2 accelerator with Gen4 PCIe using the Xilinx Kintex UltraScale+ FPGA.
The CCIX Base Specification 1.0a evaluation version allows non-member companies to explore the benefits of CCIX’s chip-to-chip interconnect. This CCIX Base Specification 1.0a evaluation version includes all of the CCIX Base Specification 1.0a chapters with the exception of the CCIX PHY chapter, which remains limited to CCIX Consortium members.
Athena is a high availability, dual-node flash storage platform that supports 24 PCIe NVMe dual ported solid state drives (SSD). Both nodes utilize a pair of the new Intel® Xeon® Scalable Processors for excellent Software Defined Storage capabilities. Coming Soon - Athena Gen2. View the video to learn more.
The SR113 offers dual usability as a rackable 4U server chassis and in a desktop tower set-up. Aiming to respond to the ever-growing demand for AI applications and server-based machine learning, this chassis supports up to 5 double-width GPGPU cards and the latest proprietary sized motherboards of leading brands.
CoolIT Systems' Direct Liquid Cooling (DLC) platform is a modular, rack-based liquid cooling solution that delivers dramatic increases in rack densities, component performance and energy efficiencies. This passive technology is a truly flexible solution that creates a TCO edge in today’s highly competitive marketplace.
The Everis PLQ2 Series provides ultra-reliable, dripless connections and disconnections that protect valuable electronics. Designed using a high-performance thermoplastic, the Everis PLQ2 connector is lightweight, chemical resistant, rated UL94 V0 flame retardant and easy to use—a perfect choice for liquid cooling of electronics applications.
Dell EMC PowerScale delivers the data needed for high-performance ML and DL. With superior capabilities for low latency, high throughput and massively parallel I/O, PowerScale is the ideal storage complement to GPU-accelerated compute for AI workloads, compressing the time needed for training and testing analytical models for multi-petabyte data sets.
Dell EMC PowerSwitch Z-series of core/aggregation switches provide optimal flexibility, performance, density and power efficiency for your data center. Offering a range of fixed-form-factor switches, the Z-Series is designed to deliver high performance for today's demanding workloads while providing headroom for future data-center demands.
Why not do your AI/ML/DL model development on a workstation instead of a cloud? Get to and through your model work blazing fast on a Dell Data Science Workstation. We'll demonstrate how fast and easy the same day setup, the data prep and model training time, and the performance is.
SMART Modular’s Gen-Z™ Micro Development Kit (μDK), is a compact standalone hardware platform designed for early adopters of the Gen-Z protocol. The μDK includes both hardware and software components for Gen-Z Host and Gen-Z Device and will run end-to-end memory load-store accesses on Gen-Z fabric out of the box.
Enterprises everywhere want to grow, to find new revenue streams, become more efficient and gain competitive edge. They’re starting to realize that the data generated by their business is not a burden, but a potent raw material. In that data is the vital insights that can transform their businesses.
Defined by data growth, converged workloads, and digital transformation, the exascale era marks the start of a new era of discovery that demands a new era of capabilities. Now you can solve your most complex problems with affordable access to supercomputing with HPE Apollo systems.
HPE Superdome Flex and HPE Superdome Flex 280 are highly modular servers that power critical applications, accelerate analytics, and tackle HPC workloads holistically. These systems accelerate discovery by addressing data-intensive workloads in one single, “shared memory” node, rather than breaking them into smaller pieces to be distributed across multiple nodes.
See the latest development in HPC storage utilizing Distributed Asynchronous Object Storage (DAOS), Intel® Optane™ and NVMe SSDs. The demo shows DAOS performing IO500 workloads in action compared side by side with sizable Lustre installation on Intel Endeavour cluster.
A collaborative effort with weather scientists enabling a better understanding of how the earth's climate change impacts weather. Intel HPC platforms with Intel® Optane™ memory are a game changer for interactive exploration and analysis of data sets, providing scientists ability to see data from numerous angles and viewpoints over time.
oneAPI’s cross-industry, open, standards-based unified programming model delivers a common developer experience across accelerator architectures. But how does that affect you, as the HPC developer, right now? Learn how through 3 interactive experiences illustrating; developer freedom, realizing all the hardware value, and developing performant code quickly and correctly.
The data center is the new unit of computing. Explore how the NVIDIA accelerated computing platform gives modern data centers the power to accelerate deep learning, machine learning, and high-performance computing (HPC) workloads through various products, servers, and platforms.
The NVIDIA Mellanox InfiniBand interconnect brings high-speed, extremely low-latency scalable solutions that incorporate in-network computing technology to perform data algorithms on the network, enabling the era of “data-centric” data centers. InfiniBand lets supercomputer, AI, and cloud data centers operate at any scale, while reducing operational costs and infrastructure complexity.
Oracle bare metal servers provide customers with isolation, visibility, and control with a dedicated server. The servers support applications that require high core counts, large amounts of memory, and high bandwidth - scaling up to 128 cores (the largest in the industry).
High performance file systems support workloads that require the ability to read and write data at extremely high throughput rates. OCI HPC File Systems (HFS) are available on Oracle Cloud Marketplace and make it easy to deploy a variety of industry leading high-performance file servers.
Oracle Cloud now delivers a commercially viable, cost-effective HPC platform that can be used across industries, enterprises, and governments. The platform delivers performance levels that rival on-premises solutions—but with cloud economics, on-demand availability, and scale that can support jobs running tens of thousands of cores simultaneously.
GPU-powered bare metal and virtual instances offer companies a high-performance computing platform for demanding applications that run sophisticated algorithms. Applications that rely on machine learning, image processing, and massively parallel high-performance computing jobs benefit from running GPUs as they allow companies to solve complex problems and innovate faster.
Penguin Computing’s software products enable a broad set of users. Whether you are an animator who needs remote access to your workstation, or a System administrator who needs to enable and manage cloud access to resources, or looking for a powerful cluster management software, there is a solution for you.
Whether you are looking for isolated or non-isolated DC-DC conversion, Power Stamp Alliance members are developing a growing range of high density, high current 48V to point-of-load (POL) DC-DC converters ('power stamps'). Direct conversion for processor core voltage rails brings a range of business and technical benefits.
Cloud native data services enable data scientists and developers to collaborate and accelerate time to value, whether data is at rest, in motion, or in action. With performance at scale, and data pipeline automation, Red Hat can help you deliver greater results, faster, and with more impact.
Red Hat is at the vanguard of open source technology. We evaluate and tinker with the new and cutting-edge technologies that appear on the horizon daily. From software-defined networking to blockchain to the IOT, these technologies represent the future of IT, and Red Hat is right there at the forefront.
Massive global change has brought hard choices. Technical limitations shouldn’t compound them. In a well-coordinated hybrid cloud, open infrastructure tools give you the flexibility to scale up when customers and teams need more, pull back when circumstances demand it, or hold steady until conditions change.
Samtec’s AcceleRate® family of products include AcceleRate® HD Ultra-Dense, Multi-Row Mezzanine Strips, AcceleRate® HP High-Performance Arrays and AcceleRate® Cable – the industry's slimmest cable system with direct attach technology and ultra-low skew twinax cable.
Samtec’s FireFly™ Micro Flyover System™ is the first interconnect system that gives a designer the flexibility of using micro footprint high-performance optical and low-cost copper interconnects interchangeably with the same connector system. FireFly™ supports up to 28 Gbps per lane in x4 and x12 configurations.
The innovative design of NovaRay® combines extreme density and extreme performance for next-gen system design. The fully shielded differential pair design and two reliable points of contact contribute to 112 Gbps PAM4 performance per channel and industry leading 1.33 Tbps/in2 performance
In the Fall of 2020 SDSC will launch its newest National Science Foundation (NSF)-funded supercomputer, Expanse. At over twice the performance of SDSC’s Comet system, Expanse supports SDSC's vision of 'Computing without Boundaries' by increasing the capacity and performance for thousands of users of batch-oriented and science gateway computing,
SM8266 16 Channel high performance PCIe Gen4 x4 NVMe controller enables customer differentiated Enterprise SSD solutions focused on Performance, Application Acceleration, Reliability and SKU Flexibility. With a complete development platform including a turnkey stack and reference design kit, the SM8266 accelerates development time-to-market for Data Center and Enterprise storage SSDs.
The compact quick disconnect DAG06 is perfect for integration in installations such as data centers or supercomputers. The automatic locking system makes the product safe and easy to use. Colored rings help to easily identify the inputs and outputs to insure a reliable, sealed, and secured connection between cooling circuits.
SUSE Linux Enterprise Server simplifies your efforts to run on the cloud. It is built for hybrid implementation and multi cloud environment, so you can run workloads anywhere – physical, virtual, on-prem or cloud. The cloud images are available - optimized and ready to run.
TYAN Transport HX TS75-B8252 server platform, powered by dual 2nd Gen AMD EPYC™ 7002 Series Processors, is optimized for virtualization, AI inference and in-memory computing workloads. Featuring the latest PCIe® Gen 4.0 technology and supporting up to 32 DIMM slots, the 2U system can deliver exceptional computing performance.