How similar are Power10 and Power11?

Some will say this release is more like Power10+ than a whole new Power11. Others will say that the processor inside is Power11, so that’s what it is. 

We need to look at the big picture, and there are no significant changes for those items. These include: 

PCIe slots are still Gen5 and Gen4

Is this a problem? Since most of the available cards are Gen4 and even Gen3, it’s likely that Gen6 slots would be wasted today. I expect Gen6 slots and cards to debut in the PowerNext generation of servers. 

Memory module locations and interfaces remain the same as Power10

This IBM-designed interface provides twice as many connections per CPU as competing processors. It also moves the DDR logic from the CPU to the memory module. This allowed Power10-based servers to upgrade from DDR4 to DDR5 without CPU changes. 

Logically, Power11 servers should be able to upgrade from DDR5 to DDR6 if the DDR6 specification becomes available soon. For the record, I’m not holding my breath on this! While some lesser systems may try out pre-spec DDR6 memory, it makes no sense for IBM to release anything that isn’t entirely up to the final specifications, and DDR6 is nowhere near completed. 

Before assuming all memory is the same, consider that most memory capacities doubled from Power10 to Power11 with the new higher-density chips on the modules. 

The most prominent feature now delivers 512GB of memory on a single module. That one feature has more than a thousand times the storage of the largest hard drive of the original AS/400! It’s also cheaper, faster, smaller and uses less power. Who knew? 

Looking inside the server

Many visible features are the same as Power10. These include the number and speed of slots, the number of sockets and the number of DIMM slots. Beyond the cabinet, nearly all the I/O cards and NVMe modules are also identical. There’s a new 200Gbps Ethernet adapter for those who feel the need for speed! 

One conspicuously absent member of the Power11 family is the very popular 41A single CPU socket server model. Less surprisingly, the S1012 half-wide 2U edge-level server, announced only last year, hasn’t been refreshed for Power11. 

Other physical system changes include removing support for external drawers that hold old-fashioned spinning disk drives. These couldn’t have been ordered with Power10, but they were at least supported on those servers. Power11 disk choices include NVMe and external storage. Would you like to see that list again? 

If you’ve absorbed everything I’ve said so far, you might be thinking, “Ho hum. It was time for IBM to release a new machine, so they added 1, got 11 and here we are.” 

If you think that way, you haven’t read far enough! 

How does Power11 outperform Power10?

Higher memory frequencies with lower latency

In addition to increased memory capacity in Power11, these servers significantly improve memory throughput using higher memory frequencies with lower latency. Memory in Power11 servers is all DDR5 with frequencies up to 4800 MHz. Compared to Power10 DDR5 memory, speeds are often 50% faster. From Power10 servers with DDR4, the DDR5 on Power11 operates as much as 80% faster! 

An enhanced processor chip

As mentioned previously, the processor chip is Power11, and while it has the same shape and number of pins as Power10, it’s enhanced! The Power11 processor is based on Power10, which is true, but there are updates. 

These include more active cores per socket, two spare cores per socket in most configurations and higher core frequencies. Despite having more cores and higher frequencies, they still achieve improved energy efficiency. The system’s energy efficiency mode can also be scheduled to maximise throughput during peak workload times but scale back energy use during periods of low utilisation. 

Reliability and zero planned downtime

Another focus IBM has for Power11 servers is improving uptime for the server line. We all know that Power servers have been the most reliable servers for running your business for years, but IBM isn’t resting on its laurels regarding reliability. With Power11, IBM aims for zero planned downtime, leveraging capabilities like Live Partition Mobility (LPM) for the increasingly rare cases when a server outage is required to perform an update. 

Automated data collection

Starting in the AS/400 days, IBM i gained the ability to “phone home” to report hardware and operating system issues. That capability has improved with every release. With Power11, IBM has automated even more data collection, not just from the operating system, but from all areas across the hardware and supporting components, so that when an issue arises, IBM will automatically have the needed troubleshooting data at its fingertips. For the customer, this means less work to report issues and less time for resolution. 

So, is it Power11 or Power10+?

I’d say you could call it either way, as long as it’s Power10 with more memory, faster memory, more cores and more GHz — thus, better overall system performance with greater efficiency! Additionally, increased reliability, quality and timely reporting, spare cores and reaching Six 9’s of availability. Quantum-safe encryption for Secure Boot and LPM offers improved security. And it wouldn’t be 2025 if the system didn’t gain more AI capabilities. 

While there may not have been any surprises in the naming, one thing remains clear: Power11 is IBM’s next-generation Power server. 

A renowned, research-led hospital invested in a multi-year migration of its applications to a hybrid cloud environment. However, the rise of COVID-19 increased demands for data collection and processing. The hospital’s team decided to repurpose its on-premises, non-production IBM Z mainframe as a dynamic data collection system that enabled efficient and cost-effective management, storage, analysis and archiving. 

Addressing healthcare data collection and capacity challenges

Due to increasing demand for data processing, the institution required additional capacity, leading to an estimated upgrade cost of over $1 million. While exploring options, the team discovered that IBM had ceased offering upgrades for its legacy Z systems. The team began to discuss alternative solutions to source hardware and support legacy mainframe systems. 

The hospital discovered Service Express when searching for a single vendor to assist with hardware and support. Relying on guidance from the Service Express team, the institution embraced an alternative hardware upgrade strategy that leveraged a pre-owned mainframe rather than purchasing new hardware or undergoing a total refresh. 

Partnership yields an integrated and cost-effective solution

Through this new partnership, the hospital integrated migration, transition and onsite installation services for the upgrade. In addition, Service Express provided IBM mainframe maintenance and Virtual Tape Library (VTL) support for its production and backup environments. 

By utilising previous-generation hardware, the hospital doubled its capacity and reduced CapEx by over $980,000 in three years. In addition, the organisation saved an additional $300,000 in OpEx by leveraging third-party maintenance (TPM) for its legacy IBM system. 

The global manufacturing industry faces numerous challenges, including significant disruptions due to supply chain interruptions, assembly plant closures and a limited workforce. Even the most resilient organisations struggle to adapt in uncertain times. 

Economic uncertainty brings unforeseen IT challenges 

A global manufacturer identified an urgent need to increase its mainframe computing power to deliver sustainable performance amid unplanned production requirements and improve its business continuity capacity. However, economic uncertainty caused budget constraints and created a barrier, along with the impending expense of new IBM hardware. 

Confronted with OEM production backorders and extended lead times, the company’s procurement team sought a solution to meet their needs while minimising IT spending. They identified Service Express as a potential partner because of its inventory of IBM mainframe hardware, technical expertise and third-party maintenance (TPM) services. 

Cost-effective hardware solution addresses capacity demands 

After understanding the company’s hardware goals, Service Express identified, configured, delivered and installed an IBM Z system that provided an additional 10,000 MIPs capacity. Rather than undergoing a costly hardware refresh, the company opted for the previous generation of equipment, resulting in over $1 million in CapEx savings. Additionally, leveraging TPM support saved the company over $300,000 in OpEx annually. 

Service Express helped sustain performance during a period of disruption, uncertainty and pressure to cut costs. Now, the manufacturer can seamlessly scale its computing power to address unexpected fluctuations in production cycles with trusted infrastructure and reliable support from Service Express. 

1. Moving to a subscription-based model, discontinuing free and perpetual licenses 

Companies looking to purchase new, renew, upgrade or receive maintenance support on perpetual licenses must transition to a subscription model. Previously, perpetual licenses were one-time purchases that fell under most company’s CapEx budgets. Moving to a subscription model may require companies to shift these costs to OpEx. Broadcom announced that customers will continue to access their perpetual licenses and receive free zero-day patches for supported versions of vSphere. 

2. Condensing the VMware portfolio to two primary solutions with options for add-ons 

• VMware Cloud Foundation (VCF) 
• VMware vSphere Foundation (VVF) 

In addition to the main bundles, add-ons and access to vSphere Standard and vSphere Essentials Plus are available. 

3. Calculating pricing per core (with a minimum of 16 cores) 

With new product offerings comes new pricing. Previously, VMware required one CPU license for up to 32 physical cores. The pricing for VCF and VVF is calculated based on the number of cores per CPU, with a minimum of 16 cores. 

4. Rolling out changes to their partner program and framework 

In February 2024, VMware’s partner programs transitioned to Broadcom’s invitation-only Advantage Partner Program. As outlined in Broadcom’s Partner FAQ, they’ll begin rolling out invitations to VMware partners, starting with resellers. Invitation timelines may vary by partner route and market. 

Broadcom’s VP of Global Partner and Commercial Sales shared updates to the partner program, including a net margin model; streamlined pricing, systems and tools; predictable partner profitability and deal protection; and increased resourcing and support. 

Ways to mitigate VMware’s rising costs 

Every company’s next step forward will depend on various factors, including current environments, utilisation, compatibility, budget, resources, etc. Below are potential considerations and strategies to reduce subscription costs or make room in your IT budget for the increase. 

Assess your current environment and utilisation 

You’re overpaying for an additional license if your servers have less than 16 cores per CPU. That said, now’s the time to identify servers that are underutilised, idle or spare hosts since you’ll be paying for each core. Depending on your situation, you may be able to upgrade, downgrade or refresh hardware to get the most out of your budget. 

Find cost savings opportunities 

We know every IT department is challenged to do more with less and increasing VMware subscription costs have only added more pressure. Several opportunities for cost reduction include delaying hardware refresh cycles by extending the life of your equipment and switching to third-party Data Centre Maintenance to address support costs. 

Evaluate your storage strategy  

Customers can utilise the vSAN that’s included with their respective bundle. However, exceeding the allotted amount could overwhelm your budget and incur additional capacity charges. Read more about calculating cores for vSAN in Broadcom’s article. 

Exploring VMware alternatives 

There’s no telling what the future holds for VMware post-Broadcom acquisition. As VMware competitors begin to target the company’s significant customer base, IT decision-makers must weigh the cost, risk and time associated with considering an alternative solution. 

“A large-scale migration from VMware’s server virtualization platform will require multiple person-years of effort, considerable one-time project costs and business risk, and have an elapsed duration of 18 to 48 months.” — Gartner 

We’ve compiled a list of the top VMware alternatives and their advantages and disadvantages based on findings from Ace Cloud Hosting and reviews on Gartner Peer Insights. 

How Service Express can guide you through the changes 

We understand this change is widespread and impacts many — if not all — of our customers. Although the waters seem a tad murky, the good news is that you don’t have to navigate them alone. Ultimately, our priority is to help our customers get the most out of their budget without compromising performance. Our team of Infrastructure Sales Consultants is equipped to weigh your options and advise on the best strategy to maximise your VMware investment. 

Should I purchase new, recertified or refurbished data centre equipment? 

Before identifying what type of data centre hardware to purchase, start by understanding the level of support you’ll need and whether your top priority is flexibility or cost savings. OEMs often include a 1–3-year warranty with the purchase of new equipment. However, the service and level of support may be inconsistent or vary across manufacturers.  

When weighing your options for networking, core licences and storage, your best investment is to buy new for the OEM warranty and access to software and code updates. If you’re considering servers, your decision will be driven by budget constraints and how quickly you need the server up and running. 

Sometimes, companies are required to buy new data centre hardware to meet performance, compliance, capacity or power requirements. If you don’t need to refresh your equipment or aren’t interested in chasing the latest and greatest technology, leveraging recertified or refurbished hardware may be in your best interest. 

What’s the difference between recertified and refurbished data centre hardware? 

• Recertified hardware is typically overstock or unused returns. Also known as “factory recertified,” this equipment arrives in sealed OEM packaging and includes an OEM warranty for support after the sale. 
• Refurbished hardware is reconfigured, retested and updated according to our stringent testing standards. While there’s no OEM warranty included, we can bundle refurbished hardware with our Data Centre Maintenance. 

How much can I expect to save by purchasing recertified or refurbished equipment? 

You can typically save 30–70% over brand-new equipment, even when purchasing from a VAR. These savings reflect the impact of depreciation. Rapid technology changes, rather than equipment quality or performance, cause the decline in hardware value. 

What’s the difference between “warranty” and “support?” 

• Warranty is not a guaranteed SLA; it’s typically parts replacement during regular business hours. 
• Support is a fixed SLA with a higher level of service; an engineer will be dispatched and come onsite. 

What are the differences between buying from an OEM, VAR or TPM provider? 

• New or recertified: OEM  
• Refurbished: You can choose between a VAR or TPM provider.  

Some products can only be sold by an authorised VAR. As a TPM provider, it’s a conflict of interest, and we’ve chosen to maintain our neutrality. There are products we want to sell, but they don’t exist in the market or haven’t matured enough. 

What do OEMs and VARs say about purchasing from TPM providers? 

If the product isn’t new or recertified, they say it came from a used entity “grey market.” There isn’t a clear definition, but they suggest that parts from third-party providers are counterfeit or not made by the OEM. It’s refurbished hardware, so this only causes confusion and uncertainty. 

Where does Service Express get its data centre hardware? 

We use two primary sources: 

1. Our IT Asset Disposition (ITAD) Team procures equipment through customer trade-ins or unused returns from a customer lease. All hardware goes through our multiple-point inspection and testing process. 
2. We purchase equipment through our long-term hardware partners. Our trusted vendors are rigorously evaluated and vetted on an ongoing basis. 

What’s included with my hardware purchase from Service Express? Does your hardware come with a warranty? 

• Recertified: If equipment is sold with the OEM warranty, the customer contacts the OEM to handle it. 
• Refurbished: In most cases, the product comes bundled with Service Express’ Data Centre Maintenance. 

We have three standard maintenance SLAs, ranging from next business day response to 24/7×365 support. While professional services aren’t included, they’re available through an additional scope of work. 

What are my leasing options? 

While we don’t lease ourselves, we have leasing partners we recommend to customers and use for many projects. We’ve noticed that several companies have recently shifted to OpEx rather than CapEx spending. 

How do you ensure there are no security risks with refurbished equipment? 

Security is our number one concern, and we use our expertise to comply with and meet required security standards. We often exceed requirements to ensure everything is wiped, factory reset and in perfect working order. The equipment we sell won’t need intellectual property from the OEM to remain secure and safe. You’ll be able to get open-source security software, updates, etc. to protect your investment. 

What steps does Service Express take for quality assurance? 

Everything funnels through our Technical Centre, where systems are built, configured, tested, packaged and shipped from the same facility. We update equipment with the most current firmware and code, run it through the manufacturer’s steps and ensure all data is completely wiped. Our standard practise is to test everything in an environment similar to yours, and our disk testing tells us the life expectancy of the hardware. Since we may also provide you service after the sale, we prioritise quality for all parts and equipment. 

What kind of lead time should I expect? 

From the time of the purchase order, two-week delivery is the norm. Next-day delivery is not uncommon for urgent needs and is an option. Depending on the project size, it could take 8–10 weeks or more for larger projects. 

What does the sale process look like? 

It all starts with a conversation. We’ll set up a call to discuss the project, conduct a needs analysis to understand the goals and bring in additional technical resources as needed. Because we’re OEM agnostic, we take a more consultative approach to offer different and customized solutions. 

We prioritize building a relationship with you while providing the best solutions. We won’t pressure you on timelines or budgets. If we can’t provide a good fit, we’ll be honest and engage a partner or direct you elsewhere for a better alternative. 

What is a tape drive? 

Tape drives store data on, you guessed it, tape operating on a sequential access model. Tape has remained a common backup method since becoming a standard for data storage in the 1950s. However, sequential access can result in slower backup and restore times, as the tape requires rewinding or fast forwarding to access specific data points. Additionally, tape drives require precise storage conditions. Slight deviations from recommended temperature and humidity levels can drastically affect storage life. 

What is a physical tape library? 

Physical tape libraries are typically comprised of multiple tape drives and dozens to even hundreds of physical tapes. Tape libraries traditionally leverage a robot to move tapes. This robot becomes a single point of failure for the library. No tapes can be moved when it fails, and backups can no longer run until the hardware is repaired. Manual intervention is often required for tasks such as tape rotation, storage management and offsite transportation, introducing the potential for human error and disruptions in data protection processes. 

What is a virtual tape library (VTL)? 

VTL is a disk-based storage appliance that appears as physical tape to backup software. By leveraging disk-based storage, VTLs offer faster access times, improved reliability, footprint reduction and seamless integration with existing backup software. With built-in features such as encryption, deduplication and scalability, VTLs provide organisations with the agility and efficiency needed to meet the demands of data-intensive environments. 

Advantages of VTLs 

VTLs present a multitude of advantages over physical tape libraries, including: 

• Seamless integration with existing backup software: VTLs ensure a smooth transition without significant overhauls in the backup infrastructure by appearing as a physical tape appliance to the existing backup software. 
• Fewer hardware failures: Leveraging disk-based storage, VTLs experience fewer hardware failures than tape libraries, minimising disruptions to backup processes and enhancing overall reliability. 
• Faster backup speeds: Thanks to disk-based architecture, organisations can quickly and efficiently complete backup operations with a VTL. 
• Built-in encryption and deduplication capabilities: VTLs provide data security and optimise storage efficiency by eliminating redundant data. 
• Increased scalability: VTLs are incredibly scalable, allowing organisations to add libraries, drives and tapes seamlessly as their storage needs grow without significant disruptions to operations. 
• Connection to cloud services: VTL enables organisations to back up critical data to the cloud with the help of Managed Backup services for added flexibility and data protection. 

Is VTL the right solution for your organisation? 

Both physical and virtual tape libraries have their place in the modern-day data centre. Physical tape still serves as a reliable solution for long-term archival storage and offline backups. But as your organisation grows and evolves, it’s critical to look at your infrastructure to ensure you’re meeting your business’s needs. VTLs address traditional tape libraries’ limitations and offer many advantages that pave the way for seamless data management. 

At Service Express, we understand the importance of finding the right data storage solution tailored to your unique needs. Our team can use its expertise to help evaluate your organisation’s requirements and determine whether migrating to a VTL is the best strategy for your business. We can guide you through every step of the process, from assessment to implementation, ensuring a seamless transition to VTL technology.