Embedded

Marvell Introduces Six-Channel CXL 3.2 Memory Expanders Marvell Technology Inc. has introduced the Structera X 3406 and 3506, second-generation CXL 3.2 memory expansion controllers designed to increase memory capacity, bandwidth, and utilization in AI servers and data center infrastructure.

Designing AI Data Center Fabrics with PCIe 6.0 High-Radix Switches Artificial intelligence is reshaping data center architecture by changing where system bottlenecks occur. For years, networking discussions focused primarily on scale-out communication between servers and racks using Ethernet and InfiniBand. Traditional enterprise and cloud applications generated predictable north-south traffic patterns that scaled effectively as long as the external network infrastructure improved.

Chiplets at the Edge, Physical Edge Computing, Engineering Cyber-Resilient SDVs: Embedded Week Insights Here’s a roundup of this week’s must-read articles. We’ll look into the latest developments on chiplets at the edge, engineering cyber-resilient software-defined vehicles, physical edge computing, the benefits of local PCB prototyping, and more. The platform combines AI coding assistance, hardware-intent verification, open-source development, and Databricks integration.

Microchip Adds Parallel Sensing to Touch Controllers Microchip Technology Inc. has launched its first turnkey touch controllers with parallel sensing for buttons, sliders, and wheels in appliances and outdoor and industrial applications. The two new additions to the capacitive-touch family offer enhanced touch sensitivity and noise immunity, enabling touch detection through thick cover materials and gloves.

Qorvo Introduces X-Band RFFE Targeting AESA Radar Designs Qorvo Inc. has introduced an X-band RF front-end solution designed for active electronically scanned array (AESA) radar systems. The architecture combines transmit drive, receive performance, and RF front-end control within the constrained space of an array lattice, reducing the number of external components required around each antenna element.

Chiplets at the Edge: How Modular SoCs Redefine Performance Per Watt The evolution of edge systems is shifting the focus from computing power alone to how efficiently each watt is transformed into useful performance. Chiplets and modular system-on-chips (SoCs) enable the combination of CPUs, accelerators, memory, and I/O based on specific needs, creating new possibilities for designing more compact, flexible, and efficient platforms.

Silicon Labs Expands AI Tools for Embedded IoT Development Silicon Labs has expanded its developer platform with new tools for AI-assisted embedded development, hardware and software validation, open-source collaboration, and edge AI lifecycle management.

From Silicon to Cloud: Engineering Cyber-Resilient Software-Defined Vehicles Vehicles have become edge computers. Today’s electric and autonomous vehicles function less like traditional automobiles and more like edge-computing nodes on wheels: dozens of electronic control units (ECUs), high-speed Ethernet backbones, wireless radios, and cloud links all coordinating in real time.

Innodisk Unveils Industrial DDR5 8000 RDIMM Innodisk Corp. has introduced an industrial-grade DDR5 8000 RDIMM designed for compute-intensive applications including edge AI, enterprise servers, and high-performance computing. The module operates at data transfer rates of up to 8000 MT/s, representing an 11.1% increase over the previous generation, and is available in capacities from 16 GB to 64 GB. The higher transfer rate meets growing memory bandwidth demands in AI and data-intensive workloads.

Why Physical Edge Computing Could Be AI’s Next Frontier Artificial intelligence is rapidly moving from the data center to the edge. Personal electronic devices, industrial controllers, cameras, vehicles, and a multitude of connected devices more frequently perform AI inference locally rather than delegating the relevant tasks to the cloud.