Thoroughly Modern: Taking Charge of Your Hardware Refresh in 2022
February 14, 2022 Peder Lundblad
IBM i shops are well known for paying a premium for their systems because they are worth it, but they are also just as well known for hanging onto aging hardware a bit past its prime, which has its own kinds of economic costs in higher maintenance and operational costs as well as the risk of a failure in some key component of the hardware or of having an older operating system release that does not have a key security vulnerability or bug patched.
IBM does a very good job of keeping hardware technology cycles regular, and so there is really no excuse for this. Continuing to rely on aging, unsupported hardware can eat away at your IT budget in unproductive ways, which lowers the ROI of the IT organization and can stifle innovation as well as leave your systems exposed to security risks.
A hardware refresh – buying a new system or upgrading to one from an older system(s) one of the best ways to improve functionality and performance – newer hardware configurations both help reduce the system’s footprint in your data center and secure your business-critical data better.
When the average lifespan of your system is approaching three to five years, it is time to consider a hardware refresh. As with everything else, a planned approach gives you the best results; out-of-cycle upgrades might be a “band aid fix,” such as adding some incremental memory or storage to boost performance, or activating a few extra cores to add compute capacity if you happen to have a Power Systems machine that has this capability – not all of them do – and the spare capacity to do so on hand. But such measures do not add any lasting value.
An effective hardware refresh is about finding the right balance of hardware, software, and services that can scale with changing business needs. This also includes evaluating the most efficient consumption models to run your workloads, which means truly considering cloud and hybrid cloud options instead of just buying another on-premises system. With IBM getting ready to launch Power10-based entry and midrange servers this spring and the high-end Power E1080 server already out and shipping in volume, one way to think about a Power10 upgrade is to make a lot of the headache of setting up a new machine someone else’s problem out there in the cloud. Then all you have to worry about is application certification and testing on the new systems software and the hardware underneath, and you shift from having to do a large capital outlay for the new system and having to amortize it over time to just paying from the operational budget, utility style, for the machine out in the cloud. And even if you do decide to stay on-premises with a new Power10 machine – or even a new Power9 machine, as some customers will do to save money by sacrificing some performance and features in the Power10 machines – you still need to have a cloud strategy.
In 2022, your hardware refresh should take into account the new and continuing strategies of the business and IT, as well as a push toward moving into a cloud environment appropriate for your business: studies indicate that companies can reduce their infrastructure costs by 30 percent to 40 percent by simply moving to cloud.
As you look to optimize your infrastructure in 2022, here are a few things you’ll want to think about:
Assessing What You Really Need
Taking a long, hard look to see if refreshing your system is really the best way forward is worth the time – does it make more sense to move some of the workload to the cloud or adopt a hybrid approach?
Some other questions you should ask as part of your assessment:
- Do we have the expertise to implement and manage infrastructure and advanced solutions (such as external storage, dual VIO configurations, global mirror)?
- What does the business require? For example, high availability – even during backups, disaster recovery, encryption, new workloads – to accommodate acquisitions, or adding QA/Dev?
- Is there a desire to change/improve backup, perhaps move away from tape and implement virtual tape?
- What options are available to me for server tiers and which one fits the needs of my business? (Scale out versus enterprise)
Hardware Management Consoles
HMCs are critical for functionalities like remote system management and Capacity on Demand (CoD) handling. You might require an HMC for your environment if you have multiple LPARs and systems.
Ultimately, if you find yourself needing an additional server(s), consider switching to the Infrastructure-as-a-Service (IaaS) model from a cloud provider. With the right fit, you will be able to reduce total cost of ownership, achieve better data redundancy and security, while also improving server scalability to meet future needs. And get your IBM i people focused on doing what they need to do, which is to modernize their applications.
If you are unsure of what you need or would like to explore options, then it’s worth your while to work with an expert who can help you build a plan and right-size your IBM i environment.
In order to size your capacity needs accurately, it’s important to look at all aspects of your data storage needs including people, technology and future business goals.
From a technical standpoint, there are many different approaches and even drive types to choose from (ranging from inexpensive HDD to high performance FCMs). Consider your everyday business requirements to determine which type is best for you.
Additionally, these are some of the advanced features that you can take advantage of:
- NVMe Flash Core Modules (FCMs): These modules offer better resiliency (by distributing data blocks based on flash blocks’ life expectancy predictions), 50 percent more space through in-chip compression without any impact on performance, and dependable performance.
- Flashcopy Services: You can choose to create readily available full data copies for specific points in time. Among other things, flashcopy not only improves the implementation of new test environments, but also makes your HA/DR backup process more efficient.
- Encryption: Advanced Encryption Standard (AES), data at rest encryption with all NVMe drives and FCMs drives, without affecting system performance
- Fibre Channel: Revisit your Fibre Channel speeds during your refresh to see what upgrades are available (64 Gb/sec is expected to be available soon) – in high-performing computing environments, higher speeds can often improve flash storage performance significantly.
Considering your next steps is an important part of closing the loop on your current hardware refresh cycle. This conversation is primarily driven by whether your team of engineers/ administrators/ operators have the capacity to efficiently manage your on-prem/ enterprise systems today. Additionally, do they have the capacity to scale to meet future needs?
Explore engagements with experts to optimize the process, thereby freeing up your resources to work on things that add value to the business. An increasing number of companies leverage IaaS and MSPs to configure, migrate and run 24x7x365 on purpose-designed infrastructure that can be scaled up (and down) as required.
These questions (and more) are part of future proofing the business, so take the time to evaluate options and map out a long-term plan.
There are many reasons and benefits to updating your hardware and optimizing your infrastructure – namely better security, workload efficiencies, opportunities for cloud and streamlined operations. It might be tempting to put off a refresh until it becomes absolutely necessary but waiting may put your business at risk and cost more in the long run. By taking a bird’s eye view of the project, you can ensure that your hardware is able to keep pace with your business as it grows, while also keeping costs in check and interruptions minimal.
It can be intimidating to approach or commit to a long-term plan without accounting for all the moving parts. Our experience with organizations across industries over the last couple of decades has proven time and again that there is no one-size-fits-all solution. At Abacus Solutions (part of the Fresche Solutions family), we have the right solutions to help you position your business for continued success.
Whether you are mapping out your refresh plan or dealing with hardware that needs imminent help, reach out to me to determine the right configuration for you, explore options, pricing and deployment models. Additionally, look out for a special session I’m hosting with my colleague Patrick Schutz on March 3, on how to right size your hardware and lower TCO and have efficient ways to run your workloads.
Coming up the ranks in the IBM pre-owned/refurbished systems side of the business in the late 1980s, Peder Lundblad began to work on, build, upgrade and perform field upgrades and installations on the early generation AS/400s, and continued growing as a technical consultant and hands on engineer in the field. In the early 2000s, with an opportunity to work with the managed hosting side of the AS/400 industry, found the passion to architect and deliver IBM i solutions for partners and end users desiring to embrace the platform beyond supporting and managing internally. Almost 30 years later, continues to be an advocate for the IBM i platform and the IBM Storage and Power Systems it runs on, evangelizing the best platform on our little planet.
This content was sponsored by Fresche Solutions.