DS-SEP26-PG10_Layout 1 23/09/2026 17:11 Page 6
inDustry news
aDvice on how to navigate ai raM shortages anD price instability now available froM Direct insight
With the ongoing pressure on memory supply, driven by exponential demand from AI infrastructure, adversely impacting developers of embedded systems, Direct Insight is offering insight and expertise to help designers and buyers navigate this challenge. The compound effect of AI-fuelled supply
pressure and geopolitical uncertainties has seen both lead times and prices rise steeply, particularly for higher-performance, higher-bandwidth DDR5 RAM, resulting in specification, design and procurement hurdles for embedded development teams. “The rapid boom of AI technology has sent
shockwaves through the semiconductor memory market,” said David Pashley, co-founder & managing director at Direct Insight. “Skyrocketing demand has ramped up pressure on high performance DDR5, and that trickles down to DDR4 and so on, with prices and lead times spiralling as inventory is bought up. Then, crucially, production is shifted from commodity DRAM capacity to higher-margin HBM (High Bandwidth Memory), which is primarily used in
high-performance computing (HPC) and AI hardware to help overcome the ‘memory wall’ bottleneck. This just makes things worse for the embedded/IoT world.” According to the company, its deep
embedded development expertise and long-standing close relationships with key embedded hardware and software partners mean it is well-placed to make the best recommendation for any given project or set of requirements, as well having the agility and know-how to help smoothly implement it. “Over the last couple of decades, the
trend with embedded hardware has been to just add more and more memory capacity and capability. Traditionally considered a cheap commodity, it’s been almost a no-brainer to just double it, then double it again – in case you need it in the future,” observes Pashley. “But the current memory squeeze, driven by unprecedented demand from the AI goldrush, is ushering in a whole new age of RAM consciousness. In this new world, the RAM
is now the most expensive part of your BOM – even more expensive than the processor!” Working with an embedded systems expert like
Direct Insight can help development teams navigate such obstacles and, at the simplest level, guide their hardware and software choices to help define a less RAM-hungry strategy.
directinsight.co.uk avnet silica announces Distribution agreeMent with leM
Under a new agreement, Avnet Silica will offer LEM’s portfolio of PCB-mounted and module- mounted current sensors, covering currents from below 1A up to 42kA, voltage sensors for applications up to 6.4kV, and Rogowski coil solutions for grid monitoring and instrumentation. Backed by Avnet Silica’s extensive technical
sales and field application engineering network, customers across the EMEA region will benefit from wider access to LEM’s sensing portfolio and technical support throughout the design-in process, helping them progress projects more quickly. LEM’s sensing solutions are designed for applications across automotive, e-mobility
and EV charging, data centres, energy, industrial automation and smart infrastructure. “Through partnering with LEM, as a leading
name in electronic sensing solutions, we are expanding our capabilities to support engineers in developing in advanced power systems,” said Thomas Foj, vice president Supplier Management, Solutions & Digitalisation EMEA, Avnet Silica. “LEM’s technologies complement our semiconductor and power discretes, portfolio as well as our system-level power expertise, including our Powering the Shift initiative. Together, we can offer customers a broader range of technologies and practical design in
support, helping them address the increasing demands for energy efficiency, system reliability and smarter power management.” Philippe Rangheard, head of distribution EMEA,
LEM, said: “This partnership enhances our ability to support customers across Europe with innovative current and voltage sensing solutions for electrification, energy management, and industrial applications. We look forward to creating new opportunities together and delivering greater value to our customers.”
www.lem.com
www.avnet-silica.com
Matlantis accelerates catalyst Discovery to aDvance Materials innovation
ENEOS Holdings Corporation is using Matlantis PFP, the company’s general-purpose machine learning potential, to accelerate the discovery of new catalyst materials with NVIDIA ALCHEMI. Materials discovery has traditionally relied on a combination of
computational modelling, laboratory experiments, and trial and error. As the number of potential materials grows, researchers need computational methods that can evaluate large candidate spaces before committing resources to physical synthesis and testing. This is especially true for oxygen evolution reaction (OER) catalysts, which are critical to hydrogen production through water electrolysis. By combining PFP with NVIDIA ALCHEMI, ENEOS HD evaluated
approximately 100 million candidate structures for OER catalysts, identifying priority candidates for synthesis and experimental validation. The effort reduced a discovery process that traditionally took years to just a few months, demonstrating how AI-powered atomistic simulation and GPU-accelerated computing can enable materials researchers to explore significantly larger candidate spaces in less time. “The development of innovative materials is becoming increasingly important to achieving a carbon-neutral society,” said Takeshi Ibuka,
1 DESIGN SOLUTIONS septeMber 2026 0
general manager, AI Innovation Department, ENEOS Holdings Corporation. “This work demonstrates how large-scale computational screening can fundamentally change the way we discover and develop materials, allowing us to explore possibilities that would have been difficult to reach through conventional approaches. ENEOS will continue to push the boundaries of materials development by applying advanced computational technologies to some of the most important challenges facing a sustainable future.” Adding to this, Daisuke Okanohara, president and CEO of Matlantis,
said: “The collaboration between ENEOS, NVIDIA and Matlantis demonstrates what is possible when industry leaders bring together deep materials expertise, advanced computing and AI. Together, we are helping move materials discovery beyond the limits of traditional approaches and opening new opportunities for innovation across industries. We look forward to building on this work with ENEOS and NVIDIA to accelerate the development of the materials needed to address some of the world’s most pressing challenges.”
https://matlantis.com/en/
www.designsolutionsmag.co.uk
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52
