Flavors extra-specs

Flavors extra-specs that are supported by PowerVC. Use the Flavors extra-specs to set minimum and maximum virtual processors, minimum and maximum memory, define whether processors must be dedicated or shared, and other activities. PowerVC uses the term compute template instead of flavor.

You can find options for various operations of extra-specs by visiting the OpenStack Compute API documentation.

Table 1. Options for flavors extra-specs
Extra spec Possible values Function
powervm:ppt_ratio

A string that specifies the ratio in the format 1:value, where value is a power of 2 in the range 64 - 4096.

The Physical Page Table (PPT) ratio. The PPT ratio is the ratio of the maximum memory to the size of the Physical Page Table. It controls the size of the page table that is used by the hypervisor when performing live partition migration. The larger the PPT, the more entries are available for use by the partition for mapping virtual addresses to physical real addresses. If the PPT is set too small and the partition is running workloads, performance in the partition can decline during live partition migration. If this ratio is too large, more memory is reserved for the PPT with no performance gain. A higher ratio reserves less memory than a lower ratio. For example, a ratio of 1:4096 reserves less memory for the PPT than a ratio of 1:64 does.

If a value other than default is selected, hosts are filtered during deploys, resizes, and migrations to exclude hosts that do not support specifying the PPT ratio. However, this filtering is not applied during remote restarts.

This setting only applies to PowerVM® managed virtual machines. It is particularly useful for SAP HANA workloads. The recommended setting for SAP HANA workloads is 1:1024.

powervm:enforce_affinity_check

true or false

This attribute applies to POWER9 hosts only. When this value is true, the affinity score for the partition on the destination host is calculated before migration. If migrating to the proposed target host would result in a lesser affinity score, the migration fails.

The recommended value for SAP HANA workloads is true. The default value is false.

powervm:proc_units

Float (divisible by 0.1 for hosts with firmware level 7.5 or earlier and 0.05 for hosts with firmware level 7.6 or later)

The wanted proc_units. The value for the attribute cannot be less than one-tenth of the value that is specified for Virtual Central Processing Units (vCPUs) for hosts with firmware level 7.5 or earlier and one 20th of the value that is specified for vCPUs for hosts with firmware level 7.6 or later. If the value is not specified during deployment, it is defaulted to vCPUs * 0.5.

Note: The attribute is allowed in IBM® Power® Virtualization Center only when powervm:dedicated_proc is set to false.
powervm:min_proc_units

Float (divisible by 0.1 for hosts with firmware level 7.5 or earlier and 0.05 for hosts with firmware level 7.6 or later)

Minimum proc_units. The minimum value for the attribute is 0.1 for hosts with firmware level 7.5 or earlier and 0.05 for hosts with firmware level 7.6 or later. The maximum value must be equal to the maximum value of powervm:proc_units. If you specify the attribute, you must also specify powervm:proc_units, powervm:max_proc_units, powervm:min_vcpu, powervm:max_vcpu, and powervm:dedicated_proc. Set the powervm:dedicated_proc to false.

The value for the attribute cannot be less than one-tenth of the value that is specified for powervm:min_vcpu for hosts with firmware level 7.5 or earlier and one-twentieth of the value that is specified for powervm:min_vcpu for hosts with firmware level 7.6 or later. If you do not specify the value of the attribute during deployment, it is defaulted to equal the value of powervm:proc_units.

powervm:max_proc_units

Float (divisible by 0.1 for hosts with firmware level 7.5 or earlier and 0.05 for hosts with firmware level 7.6 or later)

Maximum proc_units. The minimum value max_proc_units can be equal to powervm:proc_units. The maximum value for the attribute cannot be more than the value of the host for maximum allowed processors per partition. If you specify this attribute, you must also specify powervm:proc_units, powervm:min_proc_units, powervm:min_vcpu, powervm:max_vcpu, and powervm:dedicated_proc. Set the powervm:dedicated_proc to false.

The value for the attribute cannot be less than one-tenth of the value that is specified for powervm:max_vcpu for hosts with firmware level 7.5 or earlier and one-twentieth of the value that is specified for powervm:max_vcpu for hosts with firmware level 7.6 or later. If you do not specify the value of the attribute during deployment, the value is defaulted to equal the value of powervm:proc_units.

powervm:min_vcpu

Integer >0

Minimum virtual processors. The minimum resource that is required for LPAR to boot is 1. The maximum value can be equal to the value, which is set to vCPUs. If you specify the value of the attribute, you must also specify value of powervm:max_vcpu. Defaults to value set for vCPUs.

powervm:max_vcpu

Integer >0

Maximum virtual processors. The minimum value for this attribute can be equal to the value for vCPUs. The maximum value cannot be more than the value of host for maximum allowed processors per partition. If you specify the value of the attribute, you must also specify powervm:min_vcpu. Defaults to value set for vCPUs.

powervm:min_mem

Integer >=256 divisible by the LMB size of the target

Minimum memory (MB). If you do not specify the value, the value is defaulted to the value for memory_mb.

powervm:max_mem

Integer >=256 divisible by the LMB size of the target

Maximum memory (MB). If you do not specify the value, the value is defaulted to the value for memory_mb.

powervm:dedicated_proc

true or false

Use dedicated processors. The attribute defaults to false.

powervm:dedicated_sharing_ mode

A string from the following values: share_idle_procs, keep_idle_procs, 'share_idle_procs_active, 'share_idle_procs_always.

Sharing mode for dedicated processors. The attribute is supported only when powervm:dedicated_proc is set to true.
share_idle_procs
Default value. Allow processor sharing when the virtual machine is inactive.
keep_idle_procs
Never allow processor sharing.
share_idle_procs_active
Allow processor sharing when the virtual machine is active.
share_idle_procs_always
Always allow processor sharing.
powervm:uncapped

true or false

LPAR can use unused processor cycles that are beyond or exceed the wanted setting of the attribute. This attribute is supported only when powervm:dedicated_proc is set to false. When powervm:dedicated_proc is set to false, powervm:uncapped defaults to true.

powervm:processor_ compatibility
A string from the following values:
  • default
  • POWER6, POWER6+, POWER6_Enhanced, POWER6+_Enhanced
  • POWER7
  • POWER8
  • POWER9_Base, or POWER9

A processor compatibility mode is a value that is assigned to a logical partition by the hypervisor that specifies the processor environment in which the logical partition can successfully operate.

powervm:shared_weight

Int (0-255)

Shared processor weight. When powervm:dedicated_proc is set to true and powervm:uncapped is also set to true, the value of the attribute defaults to 128.

powervm:availability_priority

Int (0-255)

Availability priority. The attribute priority of the server if there is a processor failure and there are not enough resources for all servers. VIOS and i5 need to remain high priority default of 191. The value of the attribute defaults to 128.

powervm:boot_volume_type

String with higher limit of 255 characters

The storage template ID to use for virtual machine boot volume.

Note: This extra spec is deprecated and continues to work only for the deployment of single disk images.
powervm:storage_ connectivity_group

String with higher limit of 255 characters

The storage connectivity group ID that describes a group of virtual I/O servers and Fibre Channel ports that are allowed for storage connectivity by the deployed virtual machine.

powervm:image_volume_type_<image source volume UUID>

String with higher limit of 255 characters

The key contains the ID of the source image volume that is cloned during the deployment. This ID can be obtained from the block_device mapping property in the image metadata. The value is the ID of the storage template that must be used when the image volume is cloned to create the VM volume during deployment. The extra spec can be repeated multiple times, with one extra spec for each image volume during a deployment.

powervm:ame_expansion_factor

String with higher limit of 4 characters

This attribute denotes the Active Memory Expansion (AME) factor or specifies that AME is disabled on the virtual machine. Specify 0 if AME is disabled. If AME is enabled, specify a value in the range: 1-10, with up to two decimal accuracy.

powervm:secure_boot

Integer with value 0 (disable), 1 (log only), or 2 (enforce).

Secure boot is supported on POWER9 systems that runs on AIX 7.2 TL3 or later. When secure boot is enabled, the system stack layers' (firmware, operating system, and applications) signed certificates are verified on startup. If verification fails, the system is not started. Secure boot is enabled by using a compute template. When secure boot is enabled for a virtual machine, it can be migrated only to a host that supports secure boot. However, you can force the virtual machine to be restarted remotely on an unsupported host by using the REST API.

powervm:srr_capability

true or false

If the value of the simplified remote restart capability is set to true for the LPAR, you can remote restart the LPAR to supported CEC or host when the source CEC or host is down. The attribute defaults to false.

powervm:shared_proc_pool_name

String with higher limit of 14 characters

Specifies the shared processor pool to be targeted during deployment of a virtual machine.

powervm:enable_lpar_metric

true or false.

When this value is true, it allows performance information collection for the virtual machine. The attribute is supported for deploy and resize operations.

powervm:pmem_vol_<integer number>

Possible integer values are from 1 to 10.

Provide persistent memory volume details, which include the list of volumes with the attribute name that is suffixed with the integer. For example,

"powervm:pmem_vol_1":"name:pmem_vol1, size:1024, affinity:true, device:DRAM",
"powervm:pmem_vol_2":"name:pmem_vol2, size:2048, affinity:false, device:DRAM"
Notes:
  • Persistent memory is not supported for IBM i partitions.
  • Persistent memory volume details can be specified outside the flavor by using the pmem-vol attribute. You can also provide the same details in a JSON format. However, you must use only one of these methods in the request body when you deploy a virtual machine.
powervm:virtual_serial_number String with higher limit of 7 characters or a string from the auto or none values. Virtual serial number (VSN) is supported only on VSN capable systems, Power9® systems with firmware FW950 and later and Power10+ systems.
  • If the value is auto, the underlying system automatically selects and assigns a VSN to the specified server.
  • If the value is none, which is the default value, no VSN is assigned to the virtual machine.
  • If the value is a valid 7-character VSN, then that value is assigned to the virtual machine.
powervm:min_affinity_score Possible integer values in the range 0-100 Specifies the minimum affinity score that is required by the virtual machine on the POWER Hypervisor (PHYP).
powervm:min_affinity_score_action
String from the following values:
  • warn
  • fail
  • none
Specifies the action that must be taken if the minimum affinity score that is required by the virtual machine is not satisfied.
powervm:lpar_placement
Integer from the following values:
  • 2
  • 4
  • 8
Specifies the placement policy to be used by the PHYP for processor and memory allocation of the virtual machine.
Notes:
  • Irrespective of the values you choose for minimum and maximum values for memory, vCPUs, or processing units determine the range that is available for resizing an active virtual machine. To resize outside the minimum and maximum range, you need to shut off the virtual machine. For example, if minimum memory is 512, memory is 1024, and maximum memory is 2048, you can resize memory to a value between 512 and 2048. If you want to opt for memory size of 256 or 4096, you must switch off the virtual machine.
  • Active resize is not allowed for ame_expansion_factor, secure_boot, ppt_ratio, and enforce_affinity_check.
  • Decreasing the memory assigned to an active virtual machine might sometime fail with error Some of the memory to be removed is being used by the operating system. Try the resize again with a larger memory.. This error depends on the guest operating system that runs on the virtual machine. If the operating system does not allow memory to be reduced while the virtual machine is in active state, resize fails. To reduce memory in such a case, you must switch off the virtual machine.