Skip to main content

GPU OAM power capping settings (trained technician only)

Refer to the following section for more information on GPU OAM power capping. This procedure is trained technician only.

GPU OAM power capping tools​

You can set GPU OAM power capping through XCC IPMI command. See the following sections for IPMI command.

XCC and AMC firmware version

Make sure XCC and AMC firmware are of the required version. You can view XCC and AMC firmware version in XCC Web GUI.

XCC and AMC required firmware version:

  • XCC firmware version must be USX318B or above. To update XCC frimware, see Lenovo XClarity Controller portal page.

  • AMC firmware version must be 6.2.0.0 or above. AMC firmware can be updated via XCC Web GUI.

Set up GPU OAM power capping after replacing the system board​

After replacing the system board, make sure to configure GPU OAM power capping.

Reading GPU OAM power capping value​

Use IPMI command to read the GPU OAM power capping value. See following for more details.

Steps:

  1. Read user-defined GPU OAM power capping value with command line below:
    ipmitool raw 0x3a 0xce 9 3 2
    The return value would be as below:
    ipmitool raw 0x3a 0xce 9 3 2 
    0[x] [y][z]

    where the [x], [y], and [z] are the first, second, and third digits respectively of a three-digit hexadecimal number. Convert the hexadecimal number to decimal number. The decimal number is the power capping value.

    For example, the return value below shows that the GPU power capping value is 580W (converted from hexadecimal number 244).
    ipmitool raw 0x3a 0xce 9 3 2 
    02 44

    Note down the power capping value.

    If the value returned as fail, proceed to Step 2.

  2. (Skip Step 2 if power capping value was read successfully in Step 1.)

    Read default GPU OAM power capping value with command line below:
    ipmitool raw 0x3a 0x0b 0xf2 0x0 0x10 0x02
    The return value would be as below:
    ipmitool raw 0x3a 0x0b 0xf2 0x0 0x10 0x02 
    0[x] [y][z]

    where the [x], [y], and [z] are the first, second, and third digits respectively of a three-digit hexadecimal number. Convert the hexadecimal number to decimal number. The decimal number is the power capping value.

    For example, the return value below shows that the GPU power capping value is 580W (converted from hexadecimal number 244).
    ipmitool raw 0x3a 0x0b 0xf2 0x0 0x10 0x02 
    02 44

    Note down the power capping value.

Configure GPU OAM power capping with IPMI commands​

Note

  • The power capping value ranges between 300W to 600W. XCC will reject value outside of this range.

  • All four GPUs are power capped to the same wattage value.

Steps:

  1. Convert the power capping wattage value from decimal number to hexadecimal number.

    Take 580W as an example, the decimal number 580 converts to hexadecimal number is: 244

  2. Set power capping with command line below:
    ipmitool raw 0x3a 0xce 9 3 0x01 0x0[x] 0x[y][z]

    Where[x], [y], and [z] are the first, second, and third digits respectively of the converted hexadecimal number.

    For example, the command line for power capping GPU to 580W is:
    ipmitool raw 0x3a 0xce 9 3 0x01 0x02 0x44
  3. After 30 to 50 seconds, read the power capping value with the following command:
    ipmitool raw 0x3a 0xce 9 3 2
    The return value is:
    ipmitool raw 0x3a 0xce 9 3 2 
    0[x] [y][z]
    Where the [x], [y], and [z] should match the first, second, and third digits respectively of the converted hexadecimal number. For example, if the return value is as below, then the GPU is successfully power capped to 580W.
    ipmitool raw 0x3a 0xce 9 3 2 
    02 44

  4. If the response power capping value is incorrect, perform DC cycle to the system, and repeat step 2 to verify the value. If the problem persists, perform AC cycle or virtual reseat, and verify again.