NVIDIA DGX B300 System User Guide

The NVIDIA DGX B300 System User Guide is also available as a PDF.

• Introduction to NVIDIA DGX B300 System
  • Hardware Overview
    • DGX B300 Physical and Operational Specifications
    • DGX B300 Feature Summary
    • Key Enhancements Over Previous DGX Systems
    • DGX B300 Exploded Views
    • Power Specifications
      • Support for PSU Redundancy and Continuous Operation
    • DGX B300 Locking Power Cord Specification
    • Using the Locking Power Cords
    • Front Panel Connections and Controls
      • With a Bezel
      • Without a Bezel
    • Rear Panel Modules
    • Motherboard Connections and Controls
    • Motherboard Tray Components
    • GPU Tray Components
  • Network Connections, Cables, and Adapters
    • DGX B300 Port Description
    • BMC Port LEDs
    • Supported Network Cables and Adapters
  • DGX B300 System Topology
  • DGX OS Software
  • Customer Support
• Connecting to DGX B300
  • Connecting to the Console
    • Direct Connection
    • Remote Connection Through the BMC
  • SSH Connection to the OS
• First Boot Setup
  • System Setup
  • Post Setup Tasks
    • Obtaining Software Updates
    • Enabling the SRP Daemon
• Quickstart and Basic Operation
  • Installation and Configuration
  • Registration
  • Obtaining an NGC Account
  • Turning DGX B300 On and Off
    • Startup Considerations
    • Shutdown Considerations
  • Verifying Functionality - Quick Health Check
  • Running the Pre-flight Test
  • Running NGC Containers with GPU Support
    • Using Native GPU Support
    • Using the NVIDIA Container Runtime for Docker
  • Managing CPU Mitigations
    • Determining the CPU Mitigation State of the DGX System
    • Disabling CPU Mitigations
    • Re-enabling CPU Mitigations
• SBIOS Settings
  • Accessing the SBIOS Setup
  • Configuring the Boot Order
  • Configuring the Local Terminal
    • Linux
    • Windows and MacOS
  • Power on or Reboot the System
• Using the BMC
  • Connecting to the BMC
  • Overview of BMC Controls
  • Open Ports
  • Configuring a Static IP Address for the BMC
    • Configuring a BMC Static Address by Using ipmitool
    • Configuring a BMC Static IP Address by Using the System BIOS
  • Changing the BMC Login Credentials
    • Username and Password Requirements
    • Procedure
  • Using the Remote Console
  • Setting Up Active Directory, LDAP, or E-Directory
  • Configuring Platform Event Filters
  • Uploading or Generating SSL Certificates
    • Viewing the SSL Certificate
    • Generating the SSL Certificate
    • Uploading the SSL Certificate
    • Updating the SBIOS Certificate
• Security
  • User Security Measures
    • Securing the BMC Port
  • System Security Measures
    • Secure Flash of DGX B300 Firmware
    • Encryption
    • NVIDIA System Manager Security
  • Secure Data Deletion
    • Prerequisites
    • Procedure
• Redfish APIs Support
  • Supported Redfish Features
  • Connectivity Between the Host and BMC
  • Redfish Examples
    • BMC Manager
    • Firmware Update
    • BIOS Settings
    • Modifying the Boot Order Using Redfish
    • Telemetry
    • Chassis
    • SEL Logs
    • Virtual Image
    • Backing Up and Restoring BMC Configurations
      • Backing Up the BMC Configuration
      • Restoring the BMC Configuration
    • Collecting BMC Debug Data
    • Clear BIOS and Reset to Factory Defaults
    • Querying GPU Power Limit
    • Power Capping
      • Services
      • Domains
      • Custom Policies
      • PSU Policies
• Safety
  • Safety Information
  • Safety Warnings and Cautions
  • Intended Application Uses
  • Site Selection
  • Equipment Handling Practices
  • Electrical Precautions
    • Power and Electrical Warnings
    • Power Cord Warnings
  • System Access Warnings
  • Rack Mount Warnings
  • Electrostatic Discharge
  • Other Hazards
    • CALIFORNIA DEPARTMENT OF TOXIC SUBSTANCES CONTROL
    • NICKEL
    • Battery Replacement
    • Cooling and Airflow

Reference

• RAS Specifications
  • Standard SEL Error Types
  • Standard SEL Error Type Descriptions
    • Process Error (AMI BIOS)
    • SpareCore (AMI BIOS)
    • IioInternal (AMI BIOS)
    • BAD E1.S
    • UPI (AMI BIOS)
    • UPI Fail Over Error
    • IEH (AMI BIOS)
    • LeakyBucket (AMI BIOS)
    • PCIe (PAS 3.0.1.3)
    • M.2 (PAS 3.0.1.5)
    • POST (EAS 6.2.2.8 Nvidia Post)
    • Restore Default
    • Secure Boot Fail
    • Fail to Find Boot Device
    • OEM Debug Mode
    • RTC Rest (NV)
    • MCA (PAS 3.0.1.2)
  • OEM SEL Records (Types C0h–DFh)
  • OEM Error Type Descriptions (Types C0h–DFh)
    • PCIe Ext-Error (PAS 3.0.1.3)
    • Memory (PAS 3.0.1.1)
    • Device Boot Fail
    • Memory Persistent Error (0xCA)
    • Memory Transient Error (0xCB)
    • ECS
    • DIMM Mapout
    • Firmware Update Status
    • Device Missing
    • PCIe Link Spd Width dn Train
    • ADDDC Event
    • Secure Boot Failure - Device
    • Shutdown Reason (Microsoft)
    • Bugcheck Parameter (Microsoft)
    • EWL Fatal Code
  • OEM SEL Records (Types E0h–FFh)
  • OEM Error Type Descriptions (Types E0h–FFh)
    • Extension to UMCE Reporting
    • MRC EWL
    • Bus OOR
    • Missing Device Power Cycle
    • Global Reset
    • retry_rd_err_log
    • BER Limit Error
    • eDPC Error
    • Advisory Non Fatal
    • Error Counter

Resources

• Compliance
  • United States
  • United States/Canada
  • Canada
  • CE
  • Australia and New Zealand
  • Brazil
  • Japan
  • South Korea
  • China
  • Taiwan
  • Russia/Kazakhstan/Belarus
  • Israel
  • India
  • South Africa
  • Great Britain (England, Wales, and Scotland)
• Third-Party License Notices
  • Micron msecli
  • Mellanox (OFED)
• Notices
  • Notice
  • Trademarks