Appendix - BlueField Management in NIC Mode
The following diagram illustrates the architecture and connectivity for managing the BlueField in NIC mode.
See this page for a detailed description of the interfaces of BlueField-3.
See this page for a detailed description of the interfaces of BlueField-2.
The following table describes the interfaces available to manage the BlueField.
Management Interface |
Description |
Comment |
SMBus (PCIe Golden Fingers) |
Enables PLDM/NC-SI over MCTP between the BlueField and the host BMC |
Enables the host BMC to monitor and reset the BlueField |
PCIe |
PCIe interface between the BlueField and host x86 server |
Enables the host x86 to update and recover the BlueField using RShim PCIe physical function (PF) when the host is trusted Info
Unavailable while in zero-trust mode. Use the 1GbE OOB interface instead.
|
20-pin connector UART/USB/I2C with DPU BMC Data ports |
Not in use in the recommended approach |
The following are recommended approaches for BlueField management in NIC mode:
From platform BMC using NCSI over SMBus. This method enables the following capabilities:
Monitoring of the BlueField device
Reset control of the BlueField device
From host x86 during a maintenance window. This method enables the following capabilities:
BlueField and upgrade and recovery
NoteIt is required to enable host PCIe RShim for this maintenance timeslot for upgrade
The following subsections describe the recommended management methods for specific tasks on the BlueField .
BlueField Update and Recovery
The NVIDIA BlueField file format for performing software upgrade is BlueField bootstream (BFB). The BFB serves both as a comprehensive upgrade and recovery solution for the BlueField.
Host PCIe RShim is used to push the BFB to facilitate the upgrade and recovery. Please refer to this page for more information.
After an upgrade/recovery, a system power cycle may be required to apply changes.
BlueField Monitoring
BlueField monitoring may be performed from the platform BMC, using NC-SI over SMBus to obtain the following information:
BlueField temperatures (board, DDR, and ports), voltages, and link states
BlueField FRU information about NIC firmware, network interfaces, etc.
Device sensor data record (SDR), sensor threshold and events, system event logs (SEL), etc.
Request the NVIDIA Specific NC-SI OEM Commands Application Note from your NVIDIA representative for more details.
BlueField and Reset Control
The platform BMC may issue a (soft or hard) reset to the BlueField using NC-SI over SMBus.
Request the NVIDIA Specific NC-SI OEM Commands Application Note from your NVIDIA representative for more details.