Silicon Photonics In The Data Center: What A CMOS Exec Needs To Know


Silicon Photonics is changing the data center, with the biggest changes still ahead. Figure 1: Google Jupiter Network for multi-thousand Ironwood TPU clusters. Source: Google Refresher for new readers: Data centers contain hundreds or thousands of racks. For example, the Nvidia GB200 NVL72 AI compute/switch rack is about 24 inches wide, about 88 inches high and 42 inches deep. I... » read more

Chip Industry Week in Review


Intel hired ex-Qualcomm GPU guru Eric Demers for the company's high-performance GPU push, setting the stage for a three-way battle with Nvidia and AMD. The key targets for Intel and AMD will be better power efficiency and a programming model that rivals CUDA, but don't expect Nvidia to stand still. Acquisitions Texas Instruments plans to acquire Silicon Labs for ~$7.5B cash to enhance i... » read more

Securing Hardware For The Quantum Era


Key Takeaways: Quantum threats to security are already real. Adversaries are already harvesting data that will be decrypted later by quantum computers. Quantum computers capable of breaking RSA and ECC may arrive as early as next year. Asymmetric encryption algorithms like RSA and ECC will become inadequate against quantum threats, while symmetric encryption (such as AES) is less vul... » read more

Consumer And Med Tech Mushroom As Quantum Closes In


Key Takeaways: Universities and companies are making devices inspired by biology and the human senses to help with health monitoring, semiconductor materials development, human-computer interfaces, and more. When this nascent technology becomes a viable product, government regulations will be needed to ensure consumer safety in tracking or treating their body and the environment. Qua... » read more

Securing UALink: Introducing A UALinkSec Security Module


By Dana Neustadter and Vincent van der Leest Over the next decade, artificial intelligence will dominate computing, driving profound changes in both hardware and software architectures. This transformation will reshape data center design and interconnect technologies, creating new opportunities for innovation. As AI workloads scale, the need for high-speed, low-latency communication between ... » read more

Navigating Software-Defined Vehicle Development


The automotive industry is experiencing a profound transformation as vehicles shift from primarily mechanical systems to high-performance, software-driven platforms. The emergence of Software-Defined Vehicles (SDVs) is reshaping the value chain, requiring OEMs and their partners to rethink design, development, and validation strategies. This content outlines how convergence of silicon and sy... » read more

Flexible ICs, MEMS, Metal Oxides Solve Fresh Problems


Key Takeaways: Flexible ICs are durable and form-fitting, but they add manufacturing challenges to already complex processes, while printed flex sensors lack infrastructure. MEMS are finding new popularity in massively parallel systems, on one device, or in many devices distributed across a network. Metal oxide-based sensors are more scalable than those relying on photonic crystals, ... » read more

Blog Review: Feb. 4


Siemens' Tova Levy examines thermal management in 3D-IC, including why heat behaves differently in vertical stacks and how to analyze and manage thermal risk earlier and more predictably to ensure a design can meet performance, reliability, and time-to-market targets. Cadence's Reela Samuel finds that known-good-die strategies, standardized die-to-die test access, and vertical reliability mo... » read more

Why Move To 2nm?


Key Takeaways: Scaling digital logic still provides significant benefits, especially lower power. Multi-die assemblies will be the predominant approach, and most of the circuitry will not be 2nm or below. While these systems are inherently more flexible, the number and complexity of tradeoffs required for optimizing PPA/C are increasing. The rollout of 2nm process nodes and ... » read more

Chiplet Fundamentals For Engineers: eBook


Multi-die assemblies are the next phase of Moore's Law, scaling up and out  to improve performance and add flexibility into designs. By decomposing SoCs into building blocks, yield improves for the individual dies and overall performance increases because a chip is no longer bound by reticle limits. But this is much harder than it sounds. Chiplets don't just snap together like LEGOs, and so... » read more

← Older posts Newer posts →