SMI300XE Practice Questions

A.  

The red severity status identifies failed components in the log.

B.  

The blades will be turning a bright orange.

C.  

The purple status indicator in the IPMI Dashboard indicates failures.

D.  

In IPMI under Configuration, configure a new fan.

A.  

sudo rocm-csv, and sudo rocm-amd.bug.txt

B.  

sudo rockm-smc, and sudo rockm-bug.report.txt

C.  

sudo rocm-gpu, sudo rocm-baseboardgpu

D.  

sudo rocm-smi -a, and sudo rocm-smi

A.  

1 person

B.  

6 people

C.  

2 people

D.  

10 people

A.  

sum.exe -i 10.1.1.100 -p ADMIN -u ADMIN -c checkbasicinfo

B.  

sum.exe call system information

C.  

sum.exe -ip 10.1.1.100 -user ADMIN -pass ADMIN -command checksysteminfo

D.  

sum.exe -I 10.1.1.100 -u ADMIN -p ADMIN -c checkassetinfo

A.  

The AGFHC tool is used to diagnose and stress test components on the MI300X. This will help identify and validate component issues.

B.  

The AGFHC tool is used to support other platforms such as Mac OS and Windows so they can all share GPU system resources.

C.  

The AGFHC tool is used to update firmware and BIOS on the GPU to ensure that it is up to datewith the latest code.

D.  

The AGFHC tool is used for checking support phone numbers and emails for AMD GPU support staff who should be the first point of contact for any issues related to AMD GPUs.

A.  

The tray will lock to prevent it from pulling too fast, and 1 person on each side of the tray needs to press down on the lever to unlock it.

B.  

The tray will vibrate three times, alerting the user that it is released from the rail mechanism and is ready to be lifted, then start flashing a red light.

C.  

The tray will start spewing liquid everywhere.

D.  

The tray will start becoming very hot.

A.  

By the discoloration

B.  

By the serial number

C.  

By the part number

D.  

By the warning light

A.  

3

B.  

1

C.  

2

D.  

None

A.  

You suspect there are GPUs not detected. To determine which GPUs are missing, open the IPMI GUI and review the GPU component section to make sure all GPUs are present.

B.  

CPU is not installed correctly on the server. Review the OS dmesg logs for any odd CPU messages, then if located proceed to reseat the CPU.

C.  

The GPU baseboard may not be detected. You can determine this by opening the IPMI GUI and reviewing the FW section.

D.  

If more than 3 GPUs are missing, then ROCm commands will not function. We need to verify that the 3 are missing in the IPMI GUI and review the GPU component section to make sure that at least 2 or fewer GPUs are missing.