Broadcom N1400GD 400GbE OCP 3.0 Adapter

Broadcom N1400GD 400GbE OCP 3.0 Adapter combines a high-bandwidth Ethernet controller and highly optimized hardware acceleration engines to enhance network performance and server efficiency. The adapter is ideal for building highly scalable, feature-rich networking solutions in data centers. The N1400GD adapter addresses performance and service demands of hardware acceleration technologies with high throughput and advanced flow processing. The adapter supports fourth-generation RDMA over Converged Ethernet (RoCE), which delivers low latency and reduces complexity deployment at scale. The Broadcom N1400GD 400GbE OCP 3.0 Adapter offers a TruFlow™ engine to increase virtual machine density by freeing up CPU cycles. The adapter is ideal for artificial intelligence (AI), machine learning (ML), high-performance computing (HPC), cloud/enterprise data center servers, network function virtualization, and mobile edge computing.

Features

  • Complies with 400G PCIe 5.0 QSFP112-DD OCP 3.0 adapter with line-rate standards
  • Full-duplex throughput
  • Single-port 400/200/100/50/25G
  • High-speed SerDes with 100/50G PAM4 and 25G NRZ capability
  • Large-scale RoCE deployment enabled with Switch Aware Rate Adjustment (SARA) and Drop Congestion Notification (DCN)
  • Optimized performance with StrataXGS® and StrataDNX™ switches
  • Inline encryption engine enables CPU offload
  • Peer memory direct for low-latency inter-GPU communication
  • On-chip tunneling protocol processing for Geneve, VXLAN, and NVGRE
  • Secure OCP 3.0 adapter solution with silicon Root-of-Trust platform security
  • TruFlow engine line-rate offload for virtual switch processing
  • TruManage™ enhances server manageability and security
  • Multi-host allows multiple CPUs to connect to a single Ethernet adapter
  • NIC partitioning supports 16 physical functions (PFs)
  • Seamless integration with Broadcom PCIe Gen 5 PCIe switches for GPU-based AI/ML servers

Applications

  • Front-end and GPU servers
  • AI and ML
  • HPC
  • Cloud/enterprise data center servers
  • Network function virtualization
  • NVMe storage disaggregation
  • Mobile edge computing
  • Storage servers
公開: 2024-05-22 | 更新済み: 2024-05-28