GPGPU
Finden Sie Ihr Produkt

Finden Sie Ihr Produkt

  • QuantaGrid S74G-2U

    Breakthrough accelerated performance for giant-scale AI-HPC applications

    • Introducing the first gen NVIDIA® MGX™ architecture with modular infrastructure
    • Powered by NVIDIA® Grace™ Hopper™ Superchip
    • Coherent memory between CPU and GPU with NVLink®- C2C interconnect
    • Optimized for memory intensive inference and HPC performance
    • arm SystemReady
  • QuantaGrid-D43N-3U

    Optimized Accelerated Server

    • Flexible acceleration card configuration optimized for both compute and graphic intensive workloads
    • Up to 128 CPU cores with 8TB memory capacity to feed high throughput accelerator cards
    • Up to 2x HDR/200GbE networking to cluster computing
    • Easy maintenance design for minimum downtime
  • QuantaGrid D74A-7U

    Accelerated Parallel Computing Performance for the Most Extreme AI-HPC Workloads

    • Multiple-GPUs Server for HPC / AI Training ( e.g LLMs / NLP ).
    • Powered by 2x 4th AMD EPYC9004 Processors and Compatible with Next-Gen AMD EPYC™ Processor in the future.
    • Introducing NVIDIA HGX Architecture and Flexible with 8x NVIDIA H100/H200 GPUs or 8x AMD MI300X GPUs.
    • 18x SFF All-NVMe Drive Bays for GPUDirect Storage and Boot Drive.
    • 10x OCP NIC 3.0 TSFF for GPUDirect RDMA.
    • Modularized Design for Easy Serviceability.
  • QuantaGrid D54U-3U

    Endless Flexibility for Diverse Applications​

    • Powered by 5th/4th Gen Intel® Xeon® Scalable Processor
    • Powered by NVIDIA® GPUs
    • PCle 5.0 & DDR5 platform ready
    • Up to 4x DW accelerator or 8x SW accelerator
    • Support active type and passive type accelerator
    • Up to 10x PCIe 5.0 NVMe drive to speed up data-loading
    • PCle 5.0 400Gb Networking for Scale-out
    • Enhanced serviceability with tool-less, hot-plug designs
  • QuantaGrid D74H-7U

    Advanced Performance for the Most Extreme AI-HPC Workloads

    • 2x Top Bin 5th/4th Gen Intel® Xeon® Processor
    • 18x SFF All-NVMe drive bays for GPUDirect storage and boot drive
    • 10x OCP NIC 3.0 TSFF for GPUDirect RDMA
    • 8x Hopper H100 SXM5 GPU modules with HGX baseboard
  • QuantaGrid D52G-4U

    All-in-one Box Prevails over AI and HPC challenge

    • Up to 8x NVIDIA® Tesla® V100 with NVLink™ support up to 300GB/s GPU to GPU communication
    • Up to 10x dual-width 300 Watt GPU or 16x single-width 75 Watt GPU support
    • Diversify GPU topology to Conquer Any Type of Parallel Computing Workload
    • Up to 4x100Gb/s High Bandwidth RDMA-enabled Network to Scale Out with Efficiency
    • 8x NVMe Storage to Accelerate Deep Learning
of
2