Timing Your Power11 Upgrade

IBM Power consultant Jaqui Lynch lays out the factors that shops need to be considering as they plot their migration

Jaqui Lynch April 22, 2026

One of the biggest factors determining your IBM Power environment’s level of security and performance is also one of the simplest: How up to date is your hardware and software?

Once a year, it’s important to review the hardware installed in your data center, along with levels of the software installed on those servers. While it’s important to stay as current as possible, it’s mandatory to be on hardware and software that is still in service and has replacement parts that are easily obtainable.

This means paying attention to the server and HMC hardware and infrastructure, OSes, VIO servers and server and I/O ﬁrmware.

What Drives the Hardware Upgrade Decision?

Every company has different triggers that cause them to upgrade their servers. They can range from devaluation timelines and lease expirations to the need for support and rapid response which typically occurs on the latest hardware.

One driver is the improved performance and flexibility with the new hardware. The ability to have fewer cores do the same work can result in fewer licenses, which provides a better TCO. There are also a significant number of new features introduced at each new level of hardware and/or software. This can range from faster cards, CPUs and memory to new features in the OS. Key features for Power11 include AI-native architecture, zero planned downtime (99.9999% availability), enhanced security and resiliency, and performance gains. Details on all of these are in the various Redbooks from IBM.

Upgrade decisions may also be driven by predetermined replacement schedules (devaluation timelines) and the time remaining on the hardware lease. Many people use a five-year devaluation period, which may be too long for the speed of technology change today. I am seeing a three-year timeline being used more so that people can take advantage of the features and improvements coming out.

Another key driver is technology obsolescence. IBM has withdrawn all Power9 servers and no longer allows activation of unused cores in those servers. The S914, S922 and S924 are at end of service (EOS) effective Jan. 31, 2026, and will require service extensions to get hardware support. Additionally, they have announced the end of marketing (EOM) of all Power10 servers effective July 31, 2026. This means that any new servers after that date will be Power11. Clearly, it is time to prepare HMCs, VIO servers and OSes for the new hardware before it becomes an emergency.

IBM Support Dates Explained

There is a lot of confusion about IBM-published support and lifecycle dates. There are multiple terms used, so I have provided an explanation below.

General availability (GA) – The date the product was made available to the public to buy or download

End of marketing (EOM) – The date you will no longer be able to purchase the product

End of support (EOS) – The last date on which IBM will deliver standard support services for a given version or model of the product

End of service pack support (EoSPS) – The last date IBM will provide new service packs and patches for a specific OS level

Extended support – Once a product reaches EOS, IBM may choose to allow you to purchase extended support for a period of time

First, Do Your Homework

Staying current wherever possible makes life much easier. Systems tend to be more secure and perform better when they are kept current. It is also much easier to migrate to new technology when you are not having to upgrade across multiple versions to do so.

Hardware or software obsolescence is a very strong driver to upgrade. IBM offers extended support options for most hardware and software once they go EOS, but it is important to keep in mind that parts may take longer to get if there is an issue. Also, support could take longer if the people who used to support that product have been moved to more current products.

The first step is to take an inventory of all the hardware and software (including third-party software) you are running. It is important to determine if software will run on and be supported on the new hardware. Sometimes old software requires you to stay on older OS levels that won’t run on the new hardware. You can use the HMCScanner to get an inventory of the hardware and OS levels, but third-party software will need more work. If you don’t have an HMC, you will have to log in and get this information manually. I/O microcode levels will also need to be gathered manually.

Once you have the levels, you can start working on a plan. The ﬁrst step is to determine what’s currently out of service or about to be. Look for any prerequisites. For example, the Power11 servers require either a virtual HMC or a 7063-CR2 HMC. If you plan to go to Power11 then you may very well have to replace your HMCs. You should also cross-check the hardware with software levels to ensure you are running levels that will be supported. This is where the Fx Level Recommendation Tool (FLRT) or FLRT Lite can be used to ﬁgure out what needs to be updated. FLRT Lite also provides links to all the EOS dates for various hardware and software products.

The key to a successful update is to have a well-planned update strategy, and to be proactive and get ahead of the phasing out of equipment or software. It’s important to go through the readmes and incorporate their notes into the plan, and to include steps in the plan for what to do when or if things go wrong.

Servers

Whether the server is new or not, IBM requires that you have a valid hardware maintenance agreement (HWMA) to download and install ﬁrmware, plus a valid software maintenance agreement (SWMA) to download and upgrade software. You can check entitlements on the IBM Entitled Software Site under “My Entitled Software.” You’ll need the server model and serial number, e.g., 9824-22A (Power11 S22A) serial ???????. For most products you’ll need all seven characters. Additionally, you will require an HMC that will support V11 of the HMC code. This means a 7063-CR2 or a virtual HMC.

VIO servers and other LPARs – Power11 servers require minimum levels for the VIO servers and OSes. You can check these required levels in the Redbooks provided for each server.

Impediments to Getting Current

There are several factors that can get in the way of catching up to the current technology. One of these is lease or devaluation timelines being too long. If a lease does not have an upgrade option built into it, you may have to remain on the old hardware beyond the date it is supported. The same applies if your devaluation timeline goes beyond the EOS dates for the hardware or software.

A second factor may be third-party software that cannot run on newer OSes or hardware. If this is the case, then you need to do a risk analysis on what is involved in staying on older, potentially unsupported, hardware and/or software.

IBM offers AIX binary compatibility, which means code that runs on earlier Power servers and AIX levels should run on newer ones, as long as the code doesn’t write directly to device drivers. However, there is no guarantee of performance. This is due to alignment errors that may occur due to the differences in hardware. If you plan to migrate old code to new servers without recompiling it, then it is recommended that you do performance testing. This can be an impediment due to the time involved.

And money is always a factor. Technology upgrades need to be budgeted for, and that can be a challenge.

To Stay Current, Start Budgeting

System maintenance has often been a process that gets pushed to the back of the priority list. None of us want to place a service call on a critical system only to ﬁnd out it is out of support and that you have to do an emergency upgrade or pay for support. With the number of server and software withdrawals and the end of service announcements, this is a good time to do an analysis of your current status and start budgeting for and planning those critical upgrades.

In this article I have touched on the HMC, servers, VIO servers and client LPARs, but don’t forget to include storage, switches and applications like Spectrum Scale or databases in your planning.

IBM has provided a number of tools to make maintenance planning far simpler. These include HMCScanner, Fix Central, FLRT, FLRT Lite and FLRTVC. I recommend regularly running HMCScanner and then putting those levels into FLRT so you can ﬁnd out about necessary upgrades before you have an issue. If you are running POWER10 or lower servers, or if your HMC is not a 7063-CR2 or a virtual HMC running V11, then it is time to update or replace them.

Performing proactive replacements and upgrades can save signiﬁcant time and money, especially if you’re able to consolidate into fewer servers and cores.