Nvidia DGX H100

To Next Page

To Previous Page

NVIDIA DGX H100 Service Manual

2. Rotate the locking mechanism for the PCI carrier out of the way:

3. Pull the M.2 riser card from the slot:

8.3. Remove the M.2 Boot Drive Carrier 47

Main Page

Table of Contents3
Introduction7
Customer-Replaceable Components7
Recommended Tools8
Customer Support8
Running the Pre-Flight Test9
Front Fan Module Replacement11
Front Fan Module Replacement Overview11
Identifying a Failed Fan Module11
Replacing and Returning the Front Fan Module15
Power Supply Replacement17
Power Supply Replacement Overview17
Identifying the Failed Power Supply17
Preparing the Power Supply for Replacement21
Replacing the Power Supply22
Locking Power Cords23
Motherboard Tray - Opening and Closing the IO Door25
Preparing the Motherboard for Service25
Release the Motherboard26
Pull Motherboard from Chassis27
Open the Motherboard IO Door28
Close the Motherboard IO Door29
Lock the Motherboard Lid30
Insert the Motherboard30
Finalize Motherboard Closing32
Motherboard Tray - Removal and Installation33
Preparing the Motherboard for Service33
Release the Motherboard34
Pull Motherboard from Chassis35
Remove the Motherboard Tray Lid36
Close the Motherboard Tray Lid37
Insert the Motherboard Tray into the Chassis38
Insert the Motherboard40
Finalize Motherboard Closing41
Nvme Cache Drive Replacement Overview43
Identifying the Failed U.2 Nvme SSD43
Identifying the Nvme Manufacturer and Model44
Replacing the U.2 Nvme Drive45
Insert the U.2 Nvme Drive47
Next Steps48
Nvme Cache Drive Post-Installation Tasks49
Recreating the Cache RAID 0 Volume49
Returning the Nvme Drive50
Nvme Boot Drive Replacement Overview51
Identify the Failed M.2 Nvme52
Remove the M.2 Boot Drive Carrier52
Remove the M.2 Drive54
Replace the M.2 Drive56
Install the M.2 Boot Drive Carrier and Close the System57
Integrate the New Drive and Complete Installation59
Boot Drive Assembly Replacement61
Boot Drive Riser Assembly Replacement Overview61
Preparing the System for Replacement62
Remove the M.2 Boot Drive Carrier62
Install the M.2 Boot Drive Carrier and Close the System64
Re-Install the System and Complete the Procedure66
DIMM Replacement67
DIMM Replacement Overview67
Identifying the Failed DIMM68
Replacing the DIMM68
Finalize DIMM Replacement70
Network Interface Card Replacement73
Network Card Replacement Overview73
Prepare the System for Replacement73
Remove the Non-Functional Card74
Install the New Card and Close the Lock75
Finalize the Network Interface Card Replacement77
Updating the Connectx-7 Firmware79
Connectx-7 I/O Replacement81
Connectx-7 I/O Card Replacement Overview81
Prepare the System for Replacement82
Remove the I/O Card above the Connectx Card to be Replaced82
Remove the Connectx Card82
Remove an IPEX Cable83
Insert an IPEX Cable84
Install Connectx Card85
Install the I/O Card above the Connectx Card86
Power on the System and Confirm the Replacement86
Front Console Board Replacement Overview87
Motherboard Tray Battery Replacement91
Motherboard Tray Battery Replacement Overview91
Identify a Failed Battery92
Prepare the System for Replacement92
Remove the PCI Ethernet Card92
Remove the Connectx Card94
Remove an IPEX Cable95
Replace the Battery96
Insert an IPEX Cable97
Install Connectx Card98
Install the PCI Ethernet Card99
Power on the System and Confirm Replacement101
Trusted Platform Module Replacement103
Trusted Platform Module Replacement Overview103
Prepare the System for Replacement104
Replace the TPM Module104
Install OSFP Carrier Module106
Finalize TPM Replacement107
Removing and Attaching the Bezel109
Bezel Removal109
Bezel Installation110
Rack Mount Kit Replacement113
Rack Mount Kit Description113
Remove Rack Mount Kit - Front Rack114
Remove Rack Mount Kit - Rear115
Confirm Necessary Screws and Washers117
Install Cage Nuts Using Template118
Install Rack Mount Kit - Front120
Install Rack Mount Kit - Rear121
Safety123
Safety Information123
Safety Warnings and Cautions123
Intended Application Uses124
Site Selection124
Equipment Handling Practices125
Electrical Precautions125
Power and Electrical Warnings125
Power Cord Warnings126
System Access Warnings126
Rack Mount Warnings127
Electrostatic Discharge128
Other Hazards128
California Department of Toxic Substances Control128
Nickel128
Battery Replacement128
Cooling and Airflow129
Compliance131
United States131
United States/Canada131
Canada132
Australia and New Zealand133
Brazil133
Japan133
South Korea135
China136
Taiwan138
Russia/Kazakhstan/Belarus139
Israel140
India140
South Africa141
Great Britain (England, Wales, and Scotland)141
Third-Party License Notices143
Micron Msecli143
Mellanox (OFED)144
Notices145
Notice145
Trademarks146
Copyright146

Nvidia DGX H100 - Page 53

Table of Contents

Other manuals for Nvidia DGX H100

Related product manuals