Power capping in Cisco UCS lets you set limits on the maximum power each server can use, helping you manage energy efficiently across different server types in Cisco UCS Manager.

Cisco UCS Manager supports power capping on the following:

• Cisco UCS 6600 Series Fabric Interconnect

• Cisco UCS Fabric Interconnects 9108 100G (Cisco UCS X-Series Direct)

• UCS 6500 Series Fabric Interconnects

• UCS 6400 Series Fabric Interconnects

You can use Policy Driven Chassis Group Power Cap, or Manual Blade Level Power Cap methods to allocate power that applies to all of the servers in a chassis.

Note

Cisco UCSX-9508 Chassis supports Policy Driven Chassis Group Cap. When you choose to select Policy Driven Chassis Group Cap, Cisco UCS Manager calculates the power allotment for Cisco UCSX-9508 Chassis and when you choose to select Manual Blade Level Power Cap, Chassis Management Controller (CMC) calculates the power allotment for Cisco UCSX-9508 Chassis.

Cisco UCS Manager provides the following power management policies to help you allocate power to your servers:

The power policy is global and is inherited by all of the chassis' managed by the Cisco UCS Manager instance. You can add the power policy to a service profile to specify the redundancy for power supplies in all chassis' in the Cisco UCS domain. This policy is also known as the PSU policy.

For more information about power supply redundancy, see Cisco UCS 5108 Server Chassis Hardware Installation Guide.

When you select the Policy Driven Chassis Group Power Cap in the Global Cap Policy, Cisco UCS can maintain the over-subscription of servers without risking power failures. You can achieve over-subscription through a two-tier process. For example, at the chassis level, Cisco UCS divides the amount of power available among members of the power group, and at the blade level, the amount of power allotted to a chassis is divided among blades based on priority.

Each time a service profile is associated or disassociated, Cisco UCS Manager recalculates the power allotment for each blade server within the chassis. If necessary, power from lower-priority service profiles is redistributed to higher-priority service profiles.

UCS power groups cap power in less than one second to safely protect data center circuit breakers. A blade must stay at its cap for 20 seconds before the chassis power distribution is optimized. This is intentionally carried out over a slower timescale to prevent reacting to transient spikes in demand.

Note	The system reserves enough power to boot a server in each slot, even if that slot is empty. This reserved power cannot be leveraged by servers requiring more power. Blades that fail to comply with the power cap are penalized.

Cisco UCS uses the priority set in the power control policy along with the blade type and configuration to calculate the initial power allocation for each blade within a chassis. During normal operation, the active blades within a chassis can borrow power from idle blades within the same chassis. If all blades are active and reach the power cap, service profiles with higher priority power control policies take precedence over service profiles with lower priority power control policies.

Priority is ranked on a scale of 1-10, where 1 indicates the highest priority and 10 indicates lowest priority. The default priority is 5.

Starting with Cisco UCS Manager 3.2(2), chassis dynamic power rebalance mechanism is enabled by default. The mechanism continuously monitors the power usage of the blade servers and adjusts the power allocation accordingly. Chassis dynamic power rebalance mechanism operates within the overall chassis power budget set by Cisco UCS Manager, which is calculated from the available PSU power and Group power.

For mission-critical application a special priority called no-cap is also available. Setting the priority to no-cap does not guarantee that a blade server gets maximum power all the time, however, it prioritizes the blade server over other servers during the chassis dynamic power rebalance budget allocations.

Note	If all the blade servers are set with no-cap priority and all of them run high power consuming loads, then there is a chance that some of the blade servers get capped under high power usage, based on the power distribution done through dynamic balance.

Global Power Control Policy options are inherited by all the chassis managed by the Cisco UCS Manager.

Starting with Cisco UCS Manager 4.1(3), a global policy called Power Save Mode is available. It is disabled by default, meaning that all PSUs present remain active regardless of power redundancy policy selection. Enabling the policy restores the older behavior..

Starting with Cisco UCS Manager 4.1(2), the power control policy is also used for regulating fans in Cisco UCS C220 M5 and C240 M5 rack servers in acoustically-sensitive environments. The Acoustic setting for these fans is only available on these servers. On C240 SD M5 rack servers, Acoustic mode is the default mode.

Starting with Cisco UCS Manager 4.2(1), the power control policy is also used for regulating cooling in potentially high-temperature environments. This option is only available with Cisco UCS C220 M6, C240 M6, C225 M6, and C245 M6 rack servers and can be used with any fan speed option.

Starting with Cisco UCS Manager 4.3(2), a global policy called Cisco UCS X9508 Chassis Power Extended Policy. This option is only available with Cisco UCS X9508 Chassis.

Note	You must include the power control policy in a service profile and that service profile must be associated with a server for it to take effect.

A power group is a set of chassis that all draw power from the same power distribution unit (PDU). In Cisco UCS Manager, you can create power groups that include one or more chassis, then set a peak power cap in AC watts for that power grouping.

Implementing power capping at the chassis level requires the following:

• IOM, CIMC, and BIOS version 1.4 or higher

• Two Power Supply Units (PSUs)

The peak power cap is a static value that represents the maximum power available to all blade servers within a given power group. If you add or remove a blade from a power group, but do not manually modify the peak power value, the power group adjusts the peak power cap to accommodate the basic power-on requirements of all blades within that power group.

A minimum of 890 AC watts should be set for each chassis. This converts to 800 watts of DC power, which is the minimum amount of power required to power an empty chassis. To associate a half-width blade, the group cap needs to be set to 1475 AC watts. For a full-width blade, it needs to be set to 2060 AC watts.

After a chassis is added to a power group, all service profile associated with the blades in the chassis become part of that power group. Similarly, if you add a new blade to a chassis, that blade inherently becomes part of the chassis' power group.

Note	Creating a power group is not the same as creating a server pool. However, you can populate a server pool with members of the same power group by creating a power qualifier and adding it to server pool policy.

When a chassis is removed or deleted, the chassis gets removed from the power group.

UCS Manager supports explicit and implicit power groups.

• Explicit: You can create a power group, add chassis' and racks, and assign a budget for the group.

• Implicit: Ensures that the chassis is always protected by limiting the power consumption within safe limits. By default, all chassis that are not part of an explicit power group are assigned to the default group and the appropriate caps are placed. New chassis that connect to UCS Manager are added to the default power group until you move them to a different power group.

The following table describes the error messages you might encounter while assigning power budget and working with power groups.

When manual blade-level power cap is configured in the global cap policy, you can set a power cap for each blade server in a Cisco UCS domain.

Note	Cisco UCSX-9508 Chassis does not support Manual Blade Level Power Cap. When you choose to select Manual Blade Level Power Cap, Chassis Management Controller (CMC) calculates the power allotment for Cisco UCSX-9508 Chassis.

The following configuration options are available:

• Watts—You can specify the maximum amount of power that the server can consume at one time. This maximum can be any amount between 0 watts and 1300 watts.

  Note	B480 M5 systems using 256GB DIMMs must have a manual blade level cap at 1300W.

• Unbounded—No power usage limitations are imposed on the server. The server can use as much power as it requires.

If the server encounters a spike in power usage that meets or exceeds the maximum configured for the server, Cisco UCS Manager does not disconnect or shut down the server. Instead, Cisco UCS Manager reduces the power that is made available to the server. This reduction can slow down the server, including a reduction in CPU speed.

Note	If you configure the manual blade-level power cap using Equipment > Policies > Global Policies > Global Power Allocation Policy, the priority set in the Power Control Policy is no longer relevant.

Globally managing the fan speed can help in power management by applying a single policy for all B-series server fans in an enclosure, based on general cooling needs. Set the fan speed on a per-chassis basis in the Global Policies. The two options are:

• Balanced—The fan runs at a faster speed when needed, based on the heat generated by the server. When possible, the fan returns to the minimum required speed. (Default.)

• Low Power—The fan runs at the minimum speed that is required to keep the server cool.

The new option takes effect when the new selection is saved. Use Low Power to save on system power.

The Global Power Profiling Policy specifies how power allocation is applied to all of the servers in a chassis. The policy applies when you set the Global Power Allocation Policy to policy-driven-chassis-group-cap . You can set the Global Power Profiling Policy to one of the following:

• Disabled—The minimum and maximum power cap values of the blades are calculated based on the static power consumption values of each of the components.

• Enabled—The minimum and maximum power cap values of the blades are measured as part of the server discovery. These values are similar to the actual power consumption of the blades.

Note	After enabling the Global Power Profiling Policy, you must re-acknowledge the blades to obtain the minimum and maximum power cap.

The Global Power Allocation Policy allows you to specify the Policy Driven Chassis Group Power Cap or Manual Blade-level Power Cap power allocation method applied to servers in a chassis.

Cisco recommends using the default Policy Driven Chassis Group Power Cap power allocation method.

Important

Any change to the Manual Blade level Power Cap configuration results in the loss of any groups or configuration options set for the Policy Driven Chassis Group Power Cap.

Note

Cisco UCSX-9508 Chassis supports Policy Driven Chassis Group Cap only. When you choose to select Policy Driven Chassis Group Cap, Cisco UCS Manager calculates the power allotment for Cisco UCS X9508 chassis and when you choose to select Manual Blade Level Power Cap, Chassis Management Controller (CMC) calculates the power allotment for Cisco UCSX-9508 Chassis.

Note	For Cisco UCSX-9508 Chassis Allocated (W) and Measured Max. (W) will not match. The max allocated values are used to calculate the chassis-level power limit and Intelligent Fabric Modules (IFM) allocates the power based on the power limit.

Cisco UCS Manager includes a global (default) power sync policy to address power synchronization issues between the associated service profiles and the servers. You can use the power sync policy to synchronize the power state when the power state of the service profile differs from the actual power state of the server. The policy allows you to control when to synchronize the power state on the associated service profiles for the servers. The power sync policy does not affect other power-related policies.

The power synchronization policy applies to all the service profiles by default. You cannot delete the default power sync policy, but you can edit the default policy. You can create your own power sync policies and apply them to the service profiles. You can also create a power sync policy that is specific to a service profile and it always takes precedence over the default policy.

Cisco UCS Manager creates a fault on the associated service profile when the power sync policy referenced in the service profile does not exist. Cisco UCS Manager automatically clears the fault once you create a power sync policy for the specified service profile or change the reference to an existing policy in the service profile.

Power capping is supported for all the Cisco UCS C-Series servers except Cisco UCS C125 M5 Servers.

You can manage power of the blade servers in the Cisco UCS 6324 Fabric Interconnect (FI), which is used for remote offices and branch sites, and for limited server deployments. UCS Manager supports Dual Line Power Supply Unit and 110V when used with the Cisco UCS 6324 Fabric Interconnect. You can manage how you want to allocate power when using 110V power supplies, because they might not provide enough power for a fully loaded chassis. Dual power supplies is standard for both AC and DC-48V on the Cisco UCS Mini 6324.