No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>Search

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

MM910 Management Module V100R001 User Guide 24

This document provides the product description, installation and configuration methods, and common operations of the E9000 server chassis management module MM910.
Rate and give feedback :
Huawei uses machine translation combined with human proofreading to translate this document to different languages in order to help you better understand the content of this document. Note: Even the most advanced machine translation cannot match the quality of professional translators. Huawei shall not bear any responsibility for translation accuracy and it is recommended that you refer to the English document (a link for which has been provided).
PSU Management

PSU Management

PSU Meter

GUI

Function

On the Power Meter page, you can view the status, rated power, and current power of each power supply unit (PSU), namely, PSU 1 to PSU 6.

NOTE:

If a PSU is in abnormal state, you can move the mouse over the status icon to view the fault cause.

For details about the physical locations of components, see the labels on the chassis emulation panel on the Chassis page. Figure 7-25 shows an example.

Figure 7-25 Emulation panel

NOTE:

When a PSU is in hibernation state, fails to communicate with the MM910, or generates the Power supply failure alarm, the current frequency of the PSU is displayed as 0.

PSU Hibernation Settings

GUI

Function

Multiple redundant PSUs waste power when the chassis power is lower. The PSU hibernation feature sets extra PSUs to the hibernation mode to save energy based on the PSU configurations and current chassis power. The 2000 W, 2500 W, and 3000 W PSUs all support this feature under an appropriate voltage.

The feature automatically sets redundant PSUs to the hibernation mode. When the chassis power increases caused by newly-installed components or a PSU exception occurs, the management software terminates the hibernation mode and wakes up all hibernated PSUs. When the chassis power increases or decreases, the MM software compares the sum of rated power of all operating PSUs with the total power consumption of the entire chassis and determines the number of PSUs to be hibernated based on the hibernation modes. Table 1 describes the hibernation parameters on the HMM WebUI.

Table 7-75 describes the parameters.

Table 7-75 Parameter description

Parameter

Description

Hibernation

  • Enable: Enables the PSU hibernation mode. Select a hibernation mode.
  • Disable: Disables the PSU hibernation mode. It is disabled by default.

Hibernation mode

  • N+1: One backup PSU exists. For example, if a chassis is configured with six PSUs and only two PSUs need to operate to supply sufficient power to the entire chassis, the number of PSUs that can enter hibernation mode is calculated as follows: 6 - (2 + 1) = 3
  • N+N: N backup PSUs exist. For example, if a chassis is configured with six PSUs and only two PSUs need to operate to supply sufficient power to the entire chassis, the number of PSUs that can enter hibernation mode is calculated as follows: 6 - (2 + 2) = 2
NOTE:

The MM910 implements PSU hibernation based on 50% of the PSU power due to factors such as the PSU working efficiency, power conversion rate, and electrical safety.

For example, if six 3000 W PSUs are configured and the total required power is 3000 W, each PSU consumes 1500 W and only two PSUs are required. If N+N PSU hibernation mode is used, two PSUs are working and two are standby, with the remaining two PSUs in hibernation mode.

Power Capping Settings

Screenshot

Function

On the Power Capping Settings page, you can configure power capping settings for a chassis or a compute node.

If the power capping function is enabled, the system controls the running frequency of the CPU on each specified node when the real-time power reaches the preset limit. In this event, the services and performance of the nodes may be affected.

NOTE:
  • You need to set and save the power limit for a chassis before setting the power limit for a node.
  • If the chassis has the "Low power" alarm, the power capping of compute nodes cannot be immediately delivered.

Table 7-76 describes the parameters.

Table 7-76 Parameter description

Parameter

Description

Actual chassis power (W)

Specifies the real-time chassis power.

Power capping

Specifies whether to enable the chassis power capping function.

Setting method: Click . If the icon changes to , the power capping function is enabled.

Power cap value (W)

Limits the maximum chassis power.

Setting rule: The power cap of the chassis must be greater than the sum of the power caps of all compute nodes.

NOTE:

The setting of the chassis power cap effective only when the real-time chassis power is greater than or equal to 70% of the chassis power cap. If this occurs, power capping automatically starts for the compute nodes.

Slot No.

Specifies the slot number of a compute node.

Actual Power (W)

Specifies the real-time power of a compute node.

Power Capping

Specifies whether to enable the power capping function for a compute node.

Setting method: Click . If the icon changes to , the power capping function is enabled.

Power Cap Value (W)

Limits the maximum power of a compute node.

Value range: The value range varies according to the compute node model. For details, see the instructions on the WebUI.

Setting rule: The value cannot be smaller than the lower limit displayed on the WebUI.

NOTE:
  • The power cap value of a compute node can be set in any of the following ways:
    • Enable the power capping function and set the power cap value.
    • Allow the power cap value to be automatically assigned.

      Power cap value = (Chassis power capped value – Current operating power of the rear modules – Current operating power of the fans and PSUs – Power cap values set for compute nodes)/(Total number of compute nodes in the chassis – Number of compute nodes with the power cap value configured)

      Here:

      Current operating power of the rear modules + Current operating power of the fans and PSUs = Real-time power of the chassis – Real-time power of all compute nodes

  • If the power cap value of a compute node reaches the limit but the chassis power cap value does not take effect, the CPU underclocking will not be performed for the compute node and services and performance will not be affected.

Power Control

GUI

NOTE:

The screenshot is for reference only.

Function

On the Power Control page, you can perform power-on, normal power-off, cold reset or forced reset, forced power off, and send non-maskable interrupts (NMIs) for compute nodes and switch modules, and graceful reboot or forced power cycles of compute nodes. The planes of switch modules are described as follows:

  • Base: specifies the GE switching plane.
  • Fabric: specifies the 10GE switching plane.
  • FC: Specifies the FC plane.

You can perform the following power control operations:

  • Power On
  • Graceful Power Off
  • Reset
  • Forced Power Cycle: The internal process for gracefully rebooting a compute node is as follows: shutting down the operating system (OS), powering off the compute node, and then powering on the compute node.
  • NMI: This button triggers a server to generate a non-maskable interrupt. Click this button only when the OS is abnormal. Do not click this button when the server is operating properly. If you click this button, the OS cannot be normally shut down.
    NOTE:

    The CH242 does not support NMI.

  • Power Cycle
NOTE:

It takes 10 minutes to wait for the OS shutdown on a compute node during a Forced Power Cycle by default. If the OS has not been shut down after 10 minutes, the compute node is directly powered off and then powered on. The waiting time is configurable on the BMC web user interface (WebUI) of the compute node.

Power Records

When both the master and slave chassis are installed, the slave chassis does not support historical power display.

GUI

Function
NOTE:

By default, the Internet Explorer enhanced security configuration is enabled on the Windows Server series operating systems, the historical power line cannot be displayed properly. You need to add the IP address of the MM910 to the Trusted sites zone.

  1. On the menu bar of Internet Explorer, choose Tools > Internet Options.

    The Internet Options dialog box is displayed.

  2. Click the Security tab.
  3. Click Trusted sites, and click Sites.
  4. In the Trusted sites dialog box, click Add to add the IP address to the trusted site list, and click Close.
  5. Refresh the page.
  6. In the status bar of a browser, check whether the added site is in the Trusted sites zone.

The Power Records page allows you to view the power consumed by the chassis and each slot in the last week and last day.

  • The system collects statistics on the chassis power every 5 minutes and displays the information.
  • The system collects statistics on the slot power every 10 minutes and displays the information.
NOTE:

The system displays the power statistics for only the slots that are not vacant and that login users have the permission to operate.

You can perform the following power control operations:

  • Refresh: clears the historical power of the selected object, and collects statistics from now on.
  • Download: downloads the power statistics in the last week of the selected object.
Translation
Download
Updated: 2019-04-10

Document ID: EDOC1000015900

Views: 68912

Downloads: 5169

Average rating:
This Document Applies to these Products
Related Documents
Related Version
Share
Previous Next