Navigating Hardware Refresh Cycles in the Data Center

From advice on decommissioning to insights around timing, get the latest analysis on hardware refresh cycles in the data center. 

Getting the timing right on hardware refresh cycles in the data center is both a science and an art. With so many considerations to weigh, the decision on when to pull the trigger is always a tricky call for companies and their data center asset managers.

Here’s our analysis of current approaches and trends around hardware refreshes in the data center.

Major Trendlines in Hardware Refresh Cycles 

Amid the accelerated growth of cloud computing in recent years, the on-premises data center is still alive and kicking—at least for now. But cloud investment remains the focus of attention. A recent report from Synergy Research showed huge growth in cloud expenditure, with hyperscale capacity set to triple in the next six years.

As cloud adoption soars, on-premises as a share of critical IT load will shrink. But the picture is not as dire as it may look for on-prem, which, despite decreasing as percent of spend, will likely continue to grow in capacity.

Unsurprisingly perhaps, businesses continue to rely on traditional data centers for their mission critical workloads, though most also incorporate the cloud as part of a hybrid architecture. According to 2023 data center survey from the Uptime Institute, 65% of organizations kept mission critical workloads off of public clouds. Of these, 64% cited security concerns as a barrier to adoption. While the latest 2024 data center survey suggests more than half of IT workloads are now in the cloud, many businesses continue to maintain their own data centers.

All things considered, on-premises is here for the foreseeable future. This is despite the relentless (and highly successful) attempts by the cloud service providers to grow their market share.

Changing Attitudes toward Hardware Refreshes

It might be natural to think that the staying power of on-premises computing will drive more frequent hardware refreshes. Surely companies always want the latest and greatest technology to power their data centers? 

In fact, the opposite may be true. Refresh cycles are broadly lengthening rather than shortening. In Uptime’s 2022 data center survey, the most common time between refreshes was five years, compared with three years in 2015. That represented a significant shift in a relatively short space of time. According to Uptime data from 2023, it is evident that this shift has since become entrenched.

So what’s driving this lengthening effect? Uptime Institute suggests that a slowing in power usage efficiency (PUE) for data center hardware has blunted a major incentive for more frequent server refreshes. Average PUE plummeted from 2.5 in 2007 to 1.65 in 2014, but improvement has since stagnated, with an average PUE of 1.56 in 2024.

As efficiency gains slow, so do the financial motivations for replacing already expensive hardware. Moore’s Law is delivering increasingly diminished returns.

Industry Arguments for Hardware Refresh

Unsurprisingly, enterprise vendors remain eager to push hardware upgrade cycles.

Take Dell, which — in a Forrester report it commissioned in 2022 — warned of the opportunity costs of not upgrading equipment in a timely fashion. 

“Security breaches and more flexible working styles are driving companies to rethink their approaches to device management and refresh. While the majority of respondents are pursuing a traditional three or four-year refresh, many are seeing the potential benefits of moving to an accelerated refresh model,” the report authors stated at the time. 

The report found that 66% of respondents experienced an overall security risk-reduction by adopting a two-year refresh cycle. Accordingly, the authors argued that two-year refresh was optimal for performance. The benefits in terms of security, employee experience, and sustainability “should lead to a more cost-effective device lifecycle overall,” they advised. 

Hyperscaler Refresh Cycles

While enterprise vendors like Dell advocate for more frequent refresh cycles, it’s important to watch what the hyperscalers are doing to properly gauge which way the wind is blowing. 

Consider Microsoft. It currently operates more than 4 million servers (and counting) across its massive data center portfolio— a huge amount of data center equipment by any measure. Formerly on a four-year refresh cycle, Microsoft CFO Amy Hood announced in 2022 that the firm would extend the life span of its servers to six years.

Equally telling is what Microsoft is doing with its retiring server hardware. In 2020, Microsoft committed to achieving a cloud hardware reuse rate of 90% by 2030. This is through a combination of repair management, reuse, and recycling. As of 2024, the firm is within a hair’s breadth of its goal, with an 82% reuse and recycle rate for cloud hardware.

“Using machine learning, we will process servers and hardware that are being decommissioned onsite,” Microsoft president Brad Smith remarked in an August 2020 statement. “We’ll sort the pieces that can be reused and repurposed by us, our customers, or sold.” 

Firms increasingly see a shift towards data center sustainability as a sound move. One main conclusion of Uptime’s 2022 report was that sustainability pressures are growing. Of those surveyed, 63% expected data centers in their region to be subject to sustainability reporting requirements in the next five years. Then there’s the cost savings. Google parent company Alphabet saved $3 billion in 2023 after switching to a six-year lifecycle. Amazon has also jumped on the six-year lifecycle bandwagon, which it expects to save almost $1 billion in the space of a quarter.

This trend toward repair and reuse will not only further lengthen the time horizon around hardware refreshes. It will also support the concrete expansion of a secondary market for used data center equipment that has been largely peripheral to this point. Think of it as circular hardware.  

Planning your Hardware Refresh

Let’s be clear: the upsides of well-planned technology refreshes are real. Although cloud adoption is more prevalent than ever, many businesses still rely upon on-prem storage for mission critical tasks. Even within the context of lengthening refresh cycles, that hardware needs to be replaced at some point so as not to miss out on performance gains.

There’s certainly a lot to consider when undergoing a hardware refresh. But don’t let that stand in your way. It makes sense to run a cost-benefit analysis against any proposed refresh. Questions to ask include:

• What are the performance and energy savings you expect to derive from the upgrade?
• Does your refresh plan sufficiently prepare your organization for future workload demands?
• What are the labor and capital costs associated with the work?
• How will you stage the work to minimize disruption to the operation?
• What is the value residing in the IT equipment you’re replacing?
• How can you claw back some of the investment in order to offset the upfront costs of the upgrade?
• What equipment can you redeploy internally, and what can you decommission and earmark for secure remarketing? 

For hands-on support planning a hardware refresh in your data center, draw on the industry expertise of Horizon Technology and engage with a partner that knows how to make a difference on your behalf.

Related Articles

Here are five related resources to help you get a better handle around planning for server and storage refreshes: 

1. Knowing your data center hardware
2. Accessing trusted support for your refresh needs 
3. Successfully managing a data center decommissioning 
4. Understanding the benefits of factory recertified drives
5. Reducing the amount of embodied carbon in your data center.