Creating an IBM Power Virtual Server instance

Creating a Power Virtual Server workspace

To create a Power Virtual Server workspace, complete the following steps:

1. Log in to the IBM Cloud catalog with your credentials.

2. In the search box, type Power Virtual Server, and click the Power Virtual Server tile.

3. Click Create a workspace.

4. Select Location type as Client location or IBM data center.

   For Client location location types, select your satellite location from the Satellite Location list.

   For IBM data center location types, select the IBM Cloud region that is closest to your physical location from the Satellite Location list. For the list of available IBM Cloud regions, see IBM Cloud regions.

5. In the Details section, provide a name for the workspace and select a resource group.

6. Click Continue. Review the workspace details and estimated cost on the Summary page.

7. Select I agree to the Terms and conditions checkbox.

8. Click Create. You are redirected to the Workspaces page where you can select an existing workspace.

For more information about the appropriate region for your workspace, see IBM Cloud regions.

When you create a workspace by using the Terraform resource ibm_resource_instance or the CLI command ibmcloud resource service-instance-create, for the service name field use the value power-iaas for IBM Power Virtual Server in IBM data center and for the plan name field use the value power-virtual-server-group or power-virtual-server-private-group for IBM Power Virtual Server Private Cloud in Client location.

Configuring a Power Virtual Server instance

To create a VSI, you must first create a Power Virtual Server workspace and then select it from the Workspaces list. The created workspaces are listed under Workspaces in the navigation panel of the Power Virtual Server user interface. Select the workspace for which you want to create an instance. Complete the following steps to create a virtual server instance:

1. Click Virtual server instances in the navigation panel. The VSIs that are associated with the selected workspace are displayed.

   Select a workspace to display the VSIs that were provisioned. If the displayed information appears outdated, refresh the page to view the latest data. For more information, see the FAQ page What should I do when I do not see the latest information in the UI.

   Client location If an error occurs immediately after you create or open the VSI, you must delete it.

2. To create a new instance, click Create instance.

3. In the General section, enter a name for the VSI in the Instance name field.

4. Optional: Enter user tags in the User tags (optional) field.

5. Set the number of VSIs that you want to create in the Number of instances field. The default is 1, and you can create up to five VSIs at a time. If you set more than one instance, additional options are displayed.

   The total due per month is dynamically updated in the Order Summary based on your selections. You can create a cost-effective Power Virtual Server instance that satisfies your business needs.

6. Choose an existing SSH key or create one to securely connect to your Power Virtual Server.

7. Optional: Expand Advanced Configurations to set more options for your VSI.

   • Virtual server pinning: By default, the Virtual server pinning option is set to off. When you enable this option, the VSI is kept on its current host; however, downtime can occur during planned and unplanned outages. To select a pinning type, set Virtual server pinning to on, and then select Soft or Hard.

     For more information about VSI pinning, see Virtual server pinning.

   • Metadata service: By default, the Metadata service option is set to off. When you enable access to the metadata service on a VSI, you have access to the metadata and identity APIs. You can use the metadata service to access information about a VSI and access IAM-enabled services. For more information, see Configuring and managing the metadata service for Power Virtual Server.

8. Complete the Boot image fields.

   You can create or provision a VSI without an initial boot image volume. VSIs without a boot volume are helpful in high availability and disaster recovery use cases. You can create a VSI without a boot volume and then attach a cloned or replicated volume to restore the backed-up VSI.

   Select the Deploy empty virtual server instance checkbox to provision a VSI without a boot volume. For more information, see Provisioning a virtual machine without an initial boot volume.

   You can create a VSI without a boot volume for AIX, IBM i, and Linux operating systems. For Linux, an IBM-provided subscription is required. If you assign a virtual serial number (VSN) to a VSI without storage, you must assign the VSN ro the VSI before you attach the boot volume and start the VSI.

   You can set the affinity policies for storage pools. For more information, see Configuring affinity policies.

   When you select Boot image, the Power Virtual Server user interface allows you to select the boot images from a set of available stock images or from a custom image in your image catalog. Custom images are images that you can import from IBM Cloud Object Storage or create from a VSI capture.

   When you select a stock image, you must also select the storage tier and the storage pool. When you select a custom image, the new VSIs are deployed into the same storage tier and pool where the image resides. You must select a storage type for stock images. Currently, you cannot mix Tier 1 and Tier 3 storage types. For more information, see Storage tiers.

   Client location If you select a custom image from a local catalog, the VSIs are deployed on a single storage tier.

   If you select AIX as the boot image, the Power Virtual Server user interface provides you with an option to configure the VSI for Epic workload. For more information about Epic workloads, see Configuring a VSI for Epic workloads.

   If you select IBM i as the boot image, the Power Virtual Server user interface provides you with the following options:

   • Add the following licenses to your VSI:

     • IBM i Cloud Storage Solution

     • IBM i PowerHA

     • Rational Dev Studio for IBM i

       Adding a license increases the service cost. The selected licenses are injected to your VSI. You can install specific solutions on your VSI, and the licenses are automatically applied. If you want to use these licensed programs on your IBM i VSI, you must order these licenses through Power Virtual Server. You cannot use existing licenses in your VSI.

   • Select the IBM i software tier from the IBM i software tier list. To select an IBM i software tier, you must select an image with OS version 7.3 or later from the Boot image field and set the Virtual serial number (VSN) as assigned.

   Assigning a VSN is not supported on a VSI with IBM i version 7.1 or earlier.

   If you select Full Linux Subscription (FLS) images, the Power Virtual Server user interface provides you with an option to pass in user data or scripts during the first boot runtime. When you finish entering the user data for Linux images, the system does validation checks on the input. No validation checks are done for AIX and bring your own license images. For more information, see Passing user-defined scripts.

   Cloud Optical Repository (COR) is a virtual image that can be deployed and used as a Network File Server (NFS) to perform various IBM i tasks that require media. This virtual optical image includes a collection of the media necessary for various IBM i tasks, for all supported IBM i releases. With the COR image deployed, a second Power Virtual Server instance can be deployed on the same VLAN that is set up as the client and pointed to the COR (target) NFS Server Instance. For more information about COR images, see Cloud Optical Repository.

   To deploy a VSI for SAP HANA workload, select one of the following options from the Operating system list:

   • Select Linux for SAP (HANA) in the IBM provided subscription section to use the IBM provided Linux subscription.
   • Select Linux for SAP (HANA) in the Client supplied subscription section to use your own license.

   To deploy an SAP certified profile from the Standard RISE or Application Server tabs, set SAP RISE deployment to on in the *Advanced Configurations section. The SAP RISE deployment option is enabled only if you select the OS as Linux for SAP (HANA) and the machine type as IBM Power10 or later in the Profile section.

   To deploy a VSI for SAP NetWeaver workloads without an SAP certified profile, complete the following steps:

   1. In the Boot image section, select Linux for SAP (NetWeaver) from the Operating system list.
   2. Select the appropriate SAP NetWeaver image from the Image list.
   3. Select a storage tier from the Tier list.
   4. Select a storage pool from the Storage pool field.
   5. Click Done editing to continue to the Profile section.

9. Complete the Profile fields by selecting the Machine type, the number of Cores, the amount of Memory (GB), and Core type.

   The core-to-virtual core ratio is 1:1. For shared processors, fractional cores round up to the nearest whole number. For example, 1.25 cores are equal to 2 virtual cores. For more information about processor types, see What's the difference between shared capped and shared uncapped processor performance? How do they compare to dedicated processor performance?. If the machine type is S922 and the operating system is IBM i, IBM i supports a maximum of 4 cores per VSI.

   When you use an AIX stock image as the boot volume, a console session is required to set the initial root user password. Without completing this step, SSH login is disabled. For more information, see How to create a new AIX VSI with SSH keys for root login.

   You must complete the following prerequisites to assign an IBM i software tier to a Power Virtual Server instance:

   • Select an IBM i image with version 7.3 or later from the Image list under the Boot image section.
   • Select an IBM Power10 or later server type from the Machine type list.
   • Complete the following steps to assign a VSN to the instance:
     • Edit the Virtual serial number (VSN) field. The VSN summary pane appears.
     • Select Auto-assign or Select from retained VSNs to assign a VSN.

   The supported IBM i software tiers are displayed in the IBM i software tier list based on the machine type that you select. The recommended IBM i software tier is displayed in the IBM i software tier field based on the number of cores and the memory size. You can select the IBM i software tier that is displayed in the IBM i software tier field or other options from the list.

   To deploy an SAP workload, complete the following steps:

   1. Select a Machine type from the list.

      • To deploy an SAP HANA profile, you can select a machine type from the list.
      • To deploy an SAP certified profile, you must select an IBM Power10 or later machine type from the list.

   2. Select a profile to deploy an SAP HANA profile.

   3. Select a profile from the Standard RISE or Application Server tab to deploy an SAP certified profile. The Standard RISE and Application Server tabs are enabled only when SAP RISE deployment is set to on in the Advanced Configurations section.

   The SAP RISE deployment option is enabled when you select Linux for SAP (HANA) from the Operating system list and IBM Power10 or later from the Machine type list.

10. Optional: Expand Advanced Configurations to configure more settings for your VSI.

    • Specify preferred processor compatibility mode: By default, this option is set to off. To use a specific processor compatibility mode, set Specify preferred processor compatibility mode to on, and then select the processor compatibility mode from the Preferred processor compatibility mode list.

      The Virtual server instance details page of a deployed VSI displays the preferred and effective processor compatibility modes that are set for a VSI.

      The preferred processor compatibility mode is the processor mode in which you want the VSI to operate. By default, Power Virtual Server sets the preferred processor compatibility mode to the highest mode that is supported by the targeted host type for the VSI.

      The effective processor compatibility mode is the processor mode that is in use for the VSI. The physical host in which the VSI runs determines the effective processor compatibility mode.

      The effective processor compatibility mode for the VSI might not match the preferred mode that is selected. If the operating system installed in the VSI does not support the preferred processor compatibility mode, the hypervisor can set the effective mode to a lesser mode than the preferred mode. However, the hypervisor cannot set the effective mode to a higher mode than the preferred mode. For more information about processor compatibility modes, see How does the processor compatibility mode work in a VSI?.

    • Automated remote restart: Automated remote restart is enabled by default for all VSIs in the Power Virtual Server environment. This feature automatically restarts your VSI on another available host if the current host fails unexpectedly. To disable this feature, set Automated remote restart to off during VSI creation. Alternatively, you can modify the settings on the Virtual server instance details page. For more information, see Disabling automated remote restart for a VSI.

      Automated remote restart does not restart pinned VSIs. Pinning VSIs to specific hosts results in extended downtime because the recovery depends on the time that is taken to repair the failed host. To minimize downtime, ensure that VSIs are not pinned to a host and are enabled for automated remote restart. For more information about VSI pinning, see Virtual server pinning and its impacts on VSI availability.

      Automated remote restart also does not restart a VSI in a server placement group that uses an affinity policy and includes other VSIs that are hard-pinned to the host. Affinity policies require all VSIs in the group to remain on the same host. A hard-pinned VSI prevents the VSIs in the group from moving to another host.

      When you disable the automated remote restart option for a VSI, the VSI cannot restart automatically on another available host if a host fails. Ensure that your high availabilityThe ability of a service or workload to withstand failures and continue providing processing capability according to some predefined service level. (HA) and disaster recoveryThe ability of a service or workload to recover from rare, major incidents and wide-scale failures, such as service disruption. This includes a physical disaster that affects an entire region, corruption of a database, or the loss of a service contributing to a workload. The impact exceeds the ability of the high availability design to handle it. (DR) policies are configured to protect the workload after a host failure. For more information, see High availability and disaster recovery.

11. Click Continue.

12. Complete the Storage volumes fields to attach or create new volumes and associate them with the VSI.

    Expand Advanced Configurations, and then set Configure for large quantity volumes to Enabled to support more than 127 (up to 500) volumes. This setting is at a VSI-level that remains unmodifiable upon provisioning.

    You cannot create or attach volumes larger than 2047 GB on IBM i-based VSIs. However, machine types E890 and E1080 are optimized to support the attachment of a higher number of volumes on IBM i-based VSIs. For more information, see Configuring for large quantity of volumes.

13. Define your Network interfaces by adding a public network, private network, or both. Public networks are not available at all IBM data center locations. When you add an existing private network, you can choose a specific IP address or have one auto-assigned.

    When you choose to provide a specific IP address, ensure that the IP address is not listed under reserved IP.

    For an AIX VSI, network interface controllers (NICs) are assigned based on the order in which you specify them during creation. To display the information about all the network interfaces after provisioning, open the AIX console and type ifconfig -a.

14. Accept the Terms of Use and click Create instance to provision a new Power Virtual Server. To view your boot images, go to Boot images after you provision the instance.

About VSN in IBM data center

You can assign a VSN to a VSI.

Assigning a VSN is not supported on a VSI with IBM i version 7.1 or earlier.

VSN has the following characteristics:

• It is a unique seven-digit identifier.
• It is used for licensing and tracking the usage of the VSI.
• It can be associated with only one VSI.
• It can be assigned to a new or an existing VSI.
• It can be assigned to an empty VSI in IBM data center.

For more information, see Assigning the virtual serial number to a logical partition.

A VSI moves across systems with its associated VSN. Therefore, when a VSN is associated with a VSI, you do not need to pin the VSI to a host for licensing or entitlement purposes.

VSN is a unique identifier and can be assigned only to one VSI at a time. If you deploy more than one VSI assigning the same VSN, only one VSI receives the VSN, and the remaining VSIs are deployed without a VSN.

You can view the details of a VSN that is associated with a VSI on the VSI details page. You can also view the details of the VSNs associated with the VSIs for the workspace on the virtual serial numbers page. The VSNs are either in assigned or in retained state.

When you upgrade an IBM i VSI that has a VSN assigned, you must contact IBM Support to update the IBM i licenses to match the new version.

Virtual server pinning and its impacts on VSI availability

VSIs can be optionally pinned to a host. Pin VSIs only when necessary, because pinning negatively affects VSI availability during planned maintenance activities. VSI pinning restricts the movement of VSIs during the maintenance operations, host failures, and other restart events.

Reusing volume names or VSI names in Power Virtual Server

To deploy a Power Virtual Server VSI, specify any name. If you delete a VSI and want to create a new VSI with the same name, wait at least an hour after the original VSI is deleted before you use the same name.

For example, you create a VSI with the name TEST-VSI and you delete this VSI later. The name TEST-VSI is not immediately available for reuse. Before you attempt to reuse the name TEST-VSI, wait at least an hour after the VSI is deleted.

Implementing SAP NetWeaver and SAP HANA in the Power Virtual Server environment

You can deploy SAP NetWeaver on an AIX or Linux® operating system, and SAP HANA on a Linux operating system, in your Power Virtual Server environment. You must consider several SAP-specific infrastructure requirements to run SAP applications on Power Virtual Server instances. For more information, see Planning your deployment and Deploying your infrastructure.

Consider an IBM Power server E980 that is running in a multiple VSI environment with at least one SAP HANA production system. You can deploy up to sixteen VSIs per physical server with dedicated or dedicated-donating processor cores. Each concurrently running VSI must be configured according to the workload and must fulfill the SAP HANA Hardware Configuration Check Tool (HWCCT) key performance indicators (KPIs). You must also consider the minimum number of CPU cores and memory of VSIs as described in SAP Note 2188482. For more information, see SAP support Launchpad. You must have an SAP ID to access this web page.

Configuring a VSI for Epic workloads

You can configure your VSI to deploy Epic workloads when you select AIX as your operating system.

To configure a VSI for Epic workloads, select the Configure for Epic workloads checkbox on the Boot image tile. You can verify whether the deployed VSI supports Epic workloads by checking the corresponding VSI details page. On the VSI details page, the Deployment type field must be set to Epic.

In the VSI details page, for the VSIs on which Epic workloads are supported, you must not create or attach volumes from Tier 3 to avoid performance issues. For the VSIs on which Epic workloads are supported and are in a shut-down state, you must not change the core type to any value other than dedicated to avoid performance issues.

The following table describes the VSI configuration differences between workloads that support Epic and those that do not:

By default, Epic VSIs are not pinned. Unpinned Epic VSIs can be used for nonproduction workloads. For production Epic VSIs, enable pinning to avoid performance issues.

You can choose to configure a VSI for Epic workloads only when you select AIX as your operating system. The other conditions that apply are as follows:

1. Epic workloads are supported on AIX 7.2 and later. You cannot choose AIX 7.1.
2. The supported storage volume tier is Tier 1. You can change or attach a Tier 3 storage volume. However, changing the tier might lead to performance issues.
3. Supported machine types are E980 or E1080. You cannot select S922.
4. Supported core type is dedicated. You can switch to another core type, but doing so might lead to performance issues.

Configuring affinity policies

You can use the user interface to set the affinity policies for storage pools only when the total number of VSIs in your account is less than 100. If your account has more than 100 VSIs, you must use the CLI or API to set the volume affinity policies.

Select one of the following Storage pool options:

• Auto-select pool: Use this option to allow the system to automatically select a storage pool for the storage tier with sufficient capacity.

• Affinity: Use this option to identify the storage pool that must be used to place the boot volumes based on an existing VSI or storage volume from your account. The new storage volumes for the VSI are placed in the same storage pool where the affinity object resides. If you are using a PVM instance as the affinity object, the storage pool that is selected is based on the root (boot) volume of the PVM instance.

• Anti-affinity: Use this option to identify one or more storage pools that you want to exclude from getting selected to place the boot volumes. The boot volumes are placed based on one or more existing VSIs or storage volumes from your account. When you select a storage pool to create the custom image storage volumes, the storage pools in which the list of anti-affinity objects reside are not selected. If you are using VSIs as anti-affinity objects, the storage pools are excluded based on the root (boot) volume of each PVM instance that you specify.

To learn more about the flexible tier offering of Power Virtual Server, see Storage tiers.

For more information about affinity and anti-affinity policy, see What does it mean to set an affinity or anti-affinity rule?.

If you add volumes to be created and attached to your new VSI during creation, all the volumes are provisioned in the same selected storage pool. Volumes can be created in different storage pools after the VSI is provisioned.

Provisioning a virtual machine without an initial boot volume

You can create and deploy a VSI without an initial boot volume.

The VSIs without a boot volume can be used for cloning operations. These VSIs are not bootable until a boot volume is attached after provisioning. The following table shows which images are deployed based on your OS selection:

When you attach the boot volume after provisioning of the VSI, the boot image still shows the OS-specific image without the boot volume name.

When you select the Deploy empty virtual server instance checkbox, you can provision a VSI without a boot image and boot volume. Review the following table to understand how the selection of the Deploy empty virtual server instance checkbox works along with the provisioning of the large quantity of data volumes:

Configuring large quantity of data volumes on IBM data center

IBM data center

While provisioning, you can configure your VSI to enable it to attach or detach more than 127 (up to 500) data volumes from the user interface.

IBM i virtual machines in all data centers except CHE01 support the configuration of a large quantity of data volumes. Configuring the large quantity of data volumes is supported only on IBM data center.