Could the memory shortages be the making of the channel?

Partners will need to offer purchasing options, intelligent refresh and lifecycle management

Memory

The memory market continues to reel from shortages created by the diversion of resources to the manufacture of higher-margin AI infrastructure memory. DRAM (Dynamic Random-Access Memory) and NAND flash memory used in mainstream devices are now in such short supply that prices have been pushed up by 130% leading to PCs increasing by an average 21%.

While device shortages were also apparent during the COVID pandemic, prices then remained stable. Instead, today’s situation is more akin to the shortages in new car production which saw a spike in demand for second-hand cars with prices rising accordingly. OEMs are experiencing similar issues with respect to stock availability which means they are struggling to fulfil contracts, with shipments expected be down by between 6.6% to 23%, resulting in lead time slipping.

To make matters worse, geopolitical events are further compounding matters. The war in Iran is now disrupting the production of helium, which is essential for chip manufacturing, for example, with around a third of the world’s supply of helium comes from Qatar. The war is also seeing the oil needed by chipmaking nations such as South Korea and Taiwan delayed or rerouted. Consequently, there’s no real way of knowing when production might normalize.

Lack of supply chain resilience

What the shortages have revealed is a considerable lack of resiliency in the supply chain. Rigid procurement and device refresh strategies now seem woefully out of step and stockpiling inventory is not a viable option because it risks purchasing at the peak and saddling the organisation with too much stock. Instead, we need to look at how we can plan, buy and manage technology to accommodate these kinds of events.

Latest Videos From

Some organizations are playing for time by delaying device refreshes. They’re moving from a proactive three-year cycle to a “replace on failure” approach, with average life cycles then stretching to 4.3 years. But this can have a knock-on effect on efficiency and productivity, not to mention higher break-fix support costs. Others are considering second-hand devices or switching to vendors with more available stock, but they don’t necessarily have the market knowledge or the capability to put those devices through their paces.

Then there’s the software issue. The memory shortages are also forcing organizations to look at how and where they run their software and whether they can make changes to compensate for hardware shortages. For example, some are looking to load-balance servers to use capacity more effectively.

It’s these kinds of issues that Channel partners are ideally placed to help with. Their knowledge of the market can prove indispensable, as they can draw on their relationships with manufacturers to lock down shipment schedules and pricing. They can also provide flexible financing to ease budget constraints and source or even road-test equipment from alternative suppliers on behalf of their customers. That makes it more likely that organisations will end up with a ‘mixed estate’ of devices that those partners can then help to manage.

How the channel needs to adapt

But to fulfil those roles, the channel must adapt. Rather than continuing to focus on the same sourcing strategies, partners need to begin looking at where they can create efficiencies by offering advice, lifecycle management and ‘intelligent refresh’ strategies. The latter is particularly interesting because it steps away from the cyclical nature of the traditional refresh approach. Instead, it seeks to stagger and prioritise deployments by assessing device health, performance and user requirements.

Intelligent refresh uses a data-driven approach to lifecycle management. Tools such as benchmarking and Digital End User (DEX) solutions are used to assess device utilization and map personas and workloads, for example. Potential devices from different OEMs can then be tested to see if they will align with the performance requirements of the organization, and financial options explored such as CAPEX, leasing or DaaS. Once selected, the devices can be configured according to those personas, secured and asset tagged. Then DEX tools can be used to track health, usage and UX, ensuring high uptime and devices remain matched to the needs of the employees.

Taking a data-driven approach enables partners to utilize real-time performance data, which can then be used to help with planning for refresh. Because this forecasting is based on utilization, it sees blanket refresh cycles replaced with smaller strategic waves, making it easier to fulfil those orders and maintain a consistent inventory.

In addition to assisting with the sourcing of hardware, partners can also offer device lifecycle management. They can provide a range of support, from security covering patching, policy enforcement and endpoint analytics, through to rapid break-fix support such as field-based repair to keep employees productive, and device management by tracking warranties, managing spare or loaned devices and extending asset value. They can even help claw back revenue by assisting with the resale of second-hand components.

In this way, partners can step up to help their customers adapt to this new normal. The memory shortages, initially expected to last into 2027, could now last years, according to Nvidia CEO, Jensen Huang, which means organizations will have to become more strategic in how they manage, maintain, and replace their end-user devices. That will drive an uptick in demand for partners who can advise on the optimization of hardware and software to improve utilization and longevity, who can manage a mixed estate, and draw upon data to devise intelligent refresh programmes that then forecast exactly where and when new devices are needed.

Dave Gruver
CTO of End User Compute, SHI

Dave Gruver is Field CTO of End User Compute at SHI where he assists customers as they modernize, transform and grow their organizations and build stronger employee experiences.

Prior to SHI, Dave spent 26 years at AT&T focused on end user computing and deploying Windows 3.11 to call centres before he worked his way up to the role of AVP and became responsible for all end user compute operations across the company.

During his tenure he was responsible for strategy architecture and the support of over 700,000 end point devices serving over 300,000 employees and contractors.