Nvidia Bluefield-2 Dpu Intelligent Network Card Dual-Port 100gbe Connectx-6

• Customization: Available
• Type: Wired
• Bandwidth: 200g
• Model NO.: BlueField-2 DPU
• Interface: PCI Express
• Transmission Rate: 10/25/50/100 GB/S
• Application: Server
• Connecting Medium: Optical Fiber or Dac/Aoc High-Speed Direct Cable
• Motherboard Integrated: Separate Network Card
• Frequency Range: -
• Type of Network: Cloud Computing Network, Data Center Network
• Field: Server
• Physical Interface: Dual-Port Qsfp56 Optical Port (Supporting Ethernet
• Protocol Support: 100gbe, 50gbe, 25gbe Ethernet and Infiniband
• Core Processor: Integrated 8-Core 64-Bit Armv8 Cortex-A72 Processo
• Network Adapter: Connectx-6 Dx Network Adapter
• Transport Package: Carton
• Specification: FHHL, adapted to the standard server rack, with hi
• Trademark: NVIDIA
• Origin: China

Product Description

Product Description

NVIDIA BlueField-2 DPU: Redefining Host-Level Infrastructure Performance

As the core engine for next-generation data centers, the BlueField-2 DPU injects transformative acceleration, security, and efficiency capabilities into every host:
Heterogeneous Computing Architecture
Deeply integrating the NVIDIA ConnectX-6DX Smart Network Adapter's extreme connectivity (supporting 400Gb/s high-speed interfaces) with a programmable ARM core, it offloads infrastructure tasks-such as storage, networking, and security-from the host CPU via Hardware Offloading, freeing compute resources for core business operations.
Software-Defined Universal Engine
Delivers out-of-the-box hardware acceleration for scenarios including Software-Defined Storage (SDS), Intelligent Networking (SDN), Zero-Trust Security, and Cloud-Native Management. Key features include:

  NVMe-oF protocol enabling near-local latency (<100μs) for remote storage
  Hardware-level encryption engine ensuring end-to-end data security
  Programmable architecture supporting custom network functions (e.g., load balancing, QoS control)

Performance Benefits:

In high-density AI clusters or cloud data centers, a single BlueField-2 can offload 99% of network/storage tasks, freeing 8+ CPU cores per server while boosting infrastructure efficiency by 3x, empowering enterprises to build agile, secure next-generation computing platforms.

network interface  Ethernet-10/25/50/100 Gb/s dual ports, or 200 Gb/s single port > InfiniBand-EDR/HDR100 (100 Gb/s) dual ports, or HDR (200 Gb/s) single port.	AI acceleration GPUDirect, GDS (GPUDirect Storage)	PCI Express interface  8-or 16-channel PCIe Gen 4.0, PCIe switch (with up to 8 downstream ports)
network RoCE, configuration-free RoCE; Stateless unloading	Support DDR DIMM  Single DDR4 DRAM controller > 8GB/16GB onboard DDR4 > ECC error protection support	encrypt  Enable encryption
save  NVMe SNAP, VirtIO-blk, NVMe-oF acceleration, NVMe/TCP.	Arm core  Up to 8 Armv8 A72 core (64-bit) pipelines	Host management  1GbE out-of-band management ports, NC-SI, MCTP over SMBus and MCTP over PCIe.

Core Advantages of BlueField

End-to-End Security
Supports a zero-trust security architecture, ensuring encrypted protection across data centers and edge computing environments.
Elastic Storage Optimization
Leverages NVMe-oF, GPUDirect Storage, and other technologies to achieve near-local latency (<100μs) for remote storage, with built-in data compression, deduplication, and integrity verification.
High-Performance Networking Acceleration
Delivers up to 400 Gb/s Ethernet/InfiniBand connectivity, offloading network tasks to free host CPU cores for business applications and accelerating GPU workloads.
Software-Defined Infrastructure
Enables developers to rapidly build cloud-native, DPU-accelerated infrastructure services using standard APIs via the NVIDIA DOCA SDK.

Product Features

Server-Level Zero Trust: Extends security beyond data center boundaries to enforce trustless protection at the server level.
Storage Revolution: Reduces remote storage latency by 90% and increases throughput by 3x.
Network Offloading: Offloads 99% of network tasks, freeing up 8+ CPU cores for application processing.
Developer Toolchain: DOCA SDK supports C/C++/Python for fast development of accelerated services.

Company Profile

Beijing Hengzheng TC Sci-Tech Co.,Ltd. is a  trusted supplier of rovider of data communication products and information security products solutions,headquartered in Beijing, China, with over 16 years of industry experience since our founding in 2009. Products including but not

 limited to:
Network cards, Optical modules, High-speed cables, Switches, and Industrial control boards.
 
Our dedicated team (50+ employees) specializes in networking, storage, and connectivity solutions, ensuring global customers receive high-performance hardware at competitive prices, and a prompt 7*24h service to keep your business guaranteed.

Packing & Delivery

FAQ

Q:Are you trading company or manufacturer?
A:We are manufacturer.
Q:How long is your delivery time?
A:Generally it is 2-5 days if the goods are in stock. or it is 10 days if the goods are not in stock, it is according to quantity.
Q:Can you offer a better price?Any discount?
A:Yes we can give you discount.But it is based on quantity.More quantity more attactive price.
Q:What is your warranty period?
A:1.Different products have different warranties.
2.comprehensive after-sales service, including quality service, guidance services,We focus on customer needs in every
detail.Please contact us if you need our detail warranty term.
Q:What's the package?
A:Original packing. For the outside, we will pack it in an empty box to protect from damaged during transportation, and ensure youwill receive the goods in good condition.