Introduction
The NVIDIA® BlueField®-3 data processing unit (DPU) Controller is the 3rd-generation data center infrastructure-on-a-chip that enables organizations to build software-defined, hardware-accelerated IT infrastructures from cloud to core data center to edge. With 200Gb/s Ethernet or NDR200 200Gb/s InfiniBand network connectivity, BlueField-3 DPU Controller offloads, accelerates, and isolates software-defined networking, storage, security, and management functions in ways that profoundly improve data center performance, efficiency, and security. Providing powerful computing, and a broad range of programmable acceleration engines in the I/O path, BlueField-3 is perfectly positioned to address the infrastructure needs of the most demanding applications, while delivering full software backward compatibility through the NVIDIA DOCA™ software framework.
BlueField-3 DPU Controllers are highly optimized for NVMe storage systems, Network Functions Virtualization (NFV), Cloud and Machine Learning workloads. BlueField-3 integrates all the discrete components of a storage system appliance into a single chip making it the premier solution for building Just-a-Bunch-Of-Flash (JBOF) systems, All-Flash-Array and storage appliances for NVMe over Fabrics. With an integrated NVMe-oF and NVMe/TCP offload accelerator, the BlueField-3 DPU Controller has a superior performance advantage over existing JBOF systems, significantly reducing storage transaction latency, while increasing IOPs (I/O operations per second).
Motherboard |
In PCIe x16 Configuration PCIe Gen 5.0 (32GT/s) through x16 edge connector. In PCIe x16 Extension Option - Switch DSP (Data Stream Port)
|
System Power Supply |
The maximum power consumption is up to 150W and is split between the two power sources as follows:
Note
NOTE: The power cable is not included in the package. It should be part of the system wiring or ordered separately as an accessory.
|
Operating System |
The BlueField-3 DPU Controller is shipped with Ubuntu – a Linux commercial operating system – which includes the NVIDIA OFED stack (MLNX_OFED), and is capable of running all customer-based Linux applications seamlessly. For more information, please refer to the DOCA SDK documentation or NVIDIA BlueField-3 Software User Manual. |
Connectivity |
|
Prior to unpacking your DPU Controller, it is important to make sure your server meets all the system requirements listed above for a smooth installation. Be sure to inspect each piece of equipment shipped in the packing box. If anything is missing or damaged, contact your reseller.
Card Package
Cards |
1x BlueField-3 DPU Controller |
Accessories |
1x tall bracket (shipped assembled on the card) |
Accessories Kit
This is an optional accessories kit used for debugging purposes and can be ordered separately.
Kit OPN |
Contents |
MBF35-DKIT |
4-pin USB to female USB Type-A cable |
20-pin shrouded connector to USB Type-A cable |
PCIe Auxiliary Card Package
This is an optional kit.
The PCIe auxiliary kit can be purchased separately to operate selected DPU Controllers in a dual-socket server. For package contents, refer to PCIe Auxiliary Card Kit.
For more detailed information, refer to Specifications.
This section describes hardware features and capabilities. Please refer to the relevant driver and/or firmware release notes for feature availability.
Feature |
Description |
|||||||||||||||||||||||||||||
InfiniBand Architecture Specification v1.5 compliant |
The DPU Controller delivers low latency, high bandwidth, and computing efficiency for high-performance computing (HPC), artificial intelligence (AI), and hyperscale cloud data centers applications. The DPU Controller is InfiniBand Architecture Specification v1.5 compliant. InfiniBand Network Protocols and Rates:
|
|||||||||||||||||||||||||||||
Up to 200 Gigabit Ethernet |
The DPU Controller complies with the following IEEE 802.3 standards: 200GbE / 100GbE / 50GbE / 40GbE / 25GbE / 10GbE
|
|||||||||||||||||||||||||||||
On-board Memory |
|
|||||||||||||||||||||||||||||
BlueField-3 SoC |
The DPU Controller integrates x16 Armv8.2+ A78 Hercules cores (64-bit) is interconnected by a coherent mesh network, one DRAM controller, an RDMA intelligent network adapter supporting up to 400Gb/s, an embedded PCIe switch with endpoint and root complex functionality, and up to 32 lanes of PCIe Gen 5.0. |
|||||||||||||||||||||||||||||
RDMA and RDMA over Converged InfiniBand/Ethernet (RoCE) |
The DPU Controller, utilizing IBTA RDMA (Remote Data Memory Access) and RoCE (RDMA over Converged InfiniBand/Ethernet) technology, delivers low-latency and high-performance over InfiniBand/Ethernet networks. Leveraging data center bridging (DCB) capabilities as well as advanced congestion control hardware mechanisms, RoCE provides efficient low-latency RDMA services over Layer 2 and Layer 3 networks. |
|||||||||||||||||||||||||||||
Storage Acceleration |
|
|||||||||||||||||||||||||||||
NVMe-oF and NVMe/TCP Accelerations |
Non-volatile Memory Express (NVMe) over Fabrics or over TCP are protocols for communicating block storage IO requests over RDMA or TCP to transfer data between a host computer and a target solid-state storage device or system over a network. The DPU Controller may operate as a co-processor offloading specific storage tasks from the host using its powerful NVMe Offload accelerator. |
|||||||||||||||||||||||||||||
SR-IOV |
The DPU Controller SR-IOV technology provides dedicated adapter resources and guaranteed isolation and protection for virtual machines (VM) within the server. |
|||||||||||||||||||||||||||||
GPU Direct |
GPUDirect RDMA is a technology that provides a direct P2P (Peer-to-Peer) data path between the GPU Memory directly to/from the NVIDIA HCA devices. This provides a significant decrease in GPU-GPU communication latency and completely offloads the CPU, removing it from all GPU-GPU communications across the network. The DPU Controller uses high-speed DMA transfers to copy data between P2P devices resulting in more efficient system applications |
|||||||||||||||||||||||||||||
Cryptography Accelerations |
From IPsec and TLS data-in-motion inline encryption to AES-XTS block-level data-at-rest encryption and public key acceleration, the DPU Controller hardware-based accelerations offload the crypto operations and free up the CPU, reducing latency and enabling scalable crypto solutions. BlueField-3 “host-unaware” solutions may transmit and receive data, while BlueField-3 acts as a bump-in-the-wire for crypto. |
|||||||||||||||||||||||||||||
Security Accelerators |
A consolidated compute and network solution based on the DPU Controller achieves significant advantages over a centralized security server solution. Standard encryption protocols and security applications can leverage NVIDIA BlueField-3 compute capabilities and network offloads for security application solutions such as Layer4 Statefull Firewall. |
|||||||||||||||||||||||||||||
Out-of-Band Management |
The DPU Controller incorporates a 1GbE RJ45 out-of-band port that allows the network operator to establish trust boundaries in accessing the management function to apply it to network resources. It can also be used to ensure management connectivity (including the ability to determine the status of any network component) independent of the status of other in-band network components. |
|||||||||||||||||||||||||||||
BMC |
The DPU Controller incorporate local NIC BMC (Baseboard Management Controller) hardware on the board. The BMC SoC (system on a chip) can utilize either shared or dedicated NICs for remote access. The BMC node enables remote power cycling, board environment monitoring, BlueField-3 chip temperature monitoring, board power and consumption monitoring, and individual interface resets. The BMC also supports the ability to push a bootstream to BlueField-3. Having a trusted on-board BMC that is fully isolated for the host server ensures highest security for the DPU boards. |