# Ethernet

Ethernet is a local area networking standard that specifies operation across multiple speeds using common MAC and MIB specifications.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## GUITION ESP32-S31 development kit features 7-inch capacitive touch display, Ethernet port

DevFeed: [GUITION ESP32-S31 development kit features 7-inch capacitive touch display, Ethernet port](<https://devfeed.tech/articles/guition-esp32-s31-development-kit-features-7-inch-capacitive-touch-display-ethernet-port-146397.md>)

Original publisher: [Read original article](<https://www.cnx-software.com/2026/11/20/guition-esp32-s31-development-kit-features-7-inch-capacitive-touch-display-ethernet-port/>)

Author: Jean-Luc Aufranc (CNXSoft)

Published: 2026-11-20T10:44:16Z

Content type: news

Language: en

Sources: [CNX Software - Embedded Systems News](<https://devfeed.tech/sources/cnx-software-embedded-systems-news.md>)

Topics: [ESP32](<https://devfeed.tech/topics/esp32.md>), [RISC-V](<https://devfeed.tech/topics/riscv.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>)

Tags: [802-15-4](<https://devfeed.tech/tags/802-15-4.md>), [arduino](<https://devfeed.tech/tags/arduino.md>), [battery](<https://devfeed.tech/tags/battery.md>), [ble](<https://devfeed.tech/tags/ble.md>), [bluetooth](<https://devfeed.tech/tags/bluetooth.md>), [c-c-plus-plus](<https://devfeed.tech/tags/c-c-plus-plus.md>), [capacitive](<https://devfeed.tech/tags/capacitive.md>), [capacitive-touch](<https://devfeed.tech/tags/capacitive-touch.md>), [development](<https://devfeed.tech/tags/development.md>), [development-board](<https://devfeed.tech/tags/development-board.md>), [development-kit](<https://devfeed.tech/tags/development-kit.md>), [devkit](<https://devfeed.tech/tags/devkit.md>), [display](<https://devfeed.tech/tags/display.md>), [esp32](<https://devfeed.tech/tags/esp32.md>), [esp32-s31](<https://devfeed.tech/tags/esp32-s31.md>), [espressif](<https://devfeed.tech/tags/espressif.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [guition](<https://devfeed.tech/tags/guition.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [hmi](<https://devfeed.tech/tags/hmi.md>), [port](<https://devfeed.tech/tags/port.md>), [smart-speaker](<https://devfeed.tech/tags/smart-speaker.md>), [touchscreen](<https://devfeed.tech/tags/touchscreen.md>), [video](<https://devfeed.tech/tags/video.md>), [wifi-6](<https://devfeed.tech/tags/wifi-6.md>)

### AI overview

The GUITION JC1060F570C combines an ESP32-S31 with a 7-inch, 1024x600 capacitive touchscreen and a 10/100 Mbps Ethernet port, the first of the compared ESP32-S31 kits to use the SoC's native Ethernet interface. It also provides Wi-Fi 6, Bluetooth 5.4, 802.15.4, audio hardware, USB-C, MicroSD and UART, I2C and RS485 connectors. Although the SoC is said to support Gigabit Ethernet, the board implements only Fast Ethernet; the article speculates this may reduce cost and tentatively identifies the transceiver as an IP101GR. GUITION supplies Arduino and ESP-IDF demos, documentation and Windows utilities, while noting the specifications are somewhat incomplete. The kit is available with or without an enclosure; quoted prices are identified as Black Friday 2026 promotions.

### Source excerpt

After the Waveshare ESP32-S31-Touch-LCD-4.3C and VIEWE PLCM-UES31S043H800V480C-U, the GUITION JC1060F570C is another ESP32-S31 development kit with an integrated 7-inch touchscreen display. Some companies might reconsider the way they name their product... Leaving that aside, the GUITION devkit is especially interesting as it features a large 7-inch capacitive display compared to the 4.3-inch panels used in the other two models, and it's the first to make use of the native Ethernet interface from the ESP32-S31 SoC. GUITION JC1060F570C_I_W_Y specifications: Wireless module - ESP32-S31-WROOM-3-N16R16V SoC - ESP32-S31NRV16 CPU RISC-V HP (High-performance) RV32IMAFCP CPU @ 320 MHz with FPU, SIMD, etc. RISC-V LP (Low-power) MCU core Memory - 512 KB SRAM, 32 KB low-power SRAM, 16MB PSRAM Wireless - 2.4 GHz WiFi 6 (802.11ax) Bluetooth 5.4 with support for LE Audio, direction finding, Bluetooth Mesh 1.1, and Classic (BR/EDR) 802.15.4 radio for Zigbee, Thread, and Matter Storage - 16 MB flash Antenna - PCB [...] The post GUITION ESP32-S31 development kit features 7-inch capacitive touch display, Ethernet port appeared first on CNX Software - Embedded Systems News.

## ScaleWire: An End-to-End Scale-Up Ethernet Simulation Framework for LLM Systems

DevFeed: [ScaleWire: An End-to-End Scale-Up Ethernet Simulation Framework for LLM Systems](<https://devfeed.tech/articles/scalewire-an-end-to-end-scale-up-ethernet-simulation-framework-for-llm-systems-146150.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.26086>)

Author: Kai Ma, Quanfeng Lv, Bowei Dai, Jingguo Ge, Kefan Ruan, Ruoqi Li

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Networking](<https://devfeed.tech/sources/arxiv-networking.md>)

Topics: [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [Simulation](<https://devfeed.tech/topics/simulation.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>)

Tags: [accelerators](<https://devfeed.tech/tags/accelerators.md>), [balance](<https://devfeed.tech/tags/balance.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [budget](<https://devfeed.tech/tags/budget.md>), [communication](<https://devfeed.tech/tags/communication.md>), [comparison](<https://devfeed.tech/tags/comparison.md>), [completion](<https://devfeed.tech/tags/completion.md>), [constraint](<https://devfeed.tech/tags/constraint.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cross](<https://devfeed.tech/tags/cross.md>), [cs-ni](<https://devfeed.tech/tags/cs-ni.md>), [deepseek](<https://devfeed.tech/tags/deepseek.md>), [deepseek-r1](<https://devfeed.tech/tags/deepseek-r1.md>), [design](<https://devfeed.tech/tags/design.md>), [difference](<https://devfeed.tech/tags/difference.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [framework](<https://devfeed.tech/tags/framework.md>), [llm](<https://devfeed.tech/tags/llm.md>), [scale-up](<https://devfeed.tech/tags/scale-up.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [systems](<https://devfeed.tech/tags/systems.md>)

### AI overview

ScaleWire combines packet-level simulation of Ethernet network-interface mechanisms across transport, network, and link layers with analytical estimates to explore on-chip resource sizing for LLM systems. The authors report validation against hardware measurements and nine analytical checks, and evaluate six NI designs across four communication workloads. For DeepSeek-R1 dispatch, design choices produce a 4.5x difference in per-XPU storage budget; the exploration engine uses 6-19% as many simulations as exhaustive search in grid comparisons. With vLLM prefill traces and GH200-measured kernel times, it also guides resource provisioning under an inference completion-time constraint. The abstract positions the framework as a way to balance communication performance against resource cost.

### Source excerpt

arXiv:2611.26086v1 Announce Type: new Abstract: LLM training and inference rely on scale-up networks for high-bandwidth, low-latency memory-semantic communication among accelerators (XPUs). Ethernet is emerging as an open interconnect for scale-up systems. In Ethernet systems, network-interface (NI) mechanisms across transport, network, and link layers interact, jointly shaping communication performance and on-chip resource requirements. We present ScaleWire, an end-to-end scale-up Ethernet simulation framework for LLM systems. It couples packet-level simulation of cross-layer NI mechanisms with a resource exploration engine that combines analytical estimates and simulation to size on-chip resources. We validate ScaleWire against hardware measurements and nine analytical checks, and evaluate six NI designs across four LLM communication workloads. For DeepSeek-R1 dispatch, NI design choices lead to a 4.5$\times$ difference in per-XPU on-chip storage budget. In grid-comparison experiments, the exploration engine uses 6-19% of the simulations required by exhaustive search. Using vLLM prefill traces with kernel times measured on GH200 GPUs, ScaleWire further guides resource provisioning under a completion-time constraint for inference execution. ScaleWire thus helps designers select NI mechanisms and provision on-chip resources to balance communication performance and resource cost.

## Cornelis CN6000 Doubles Omni-Path to 800G With a PCIe 6.0 SuperNIC and 48-Port Switch

DevFeed: [Cornelis CN6000 Doubles Omni-Path to 800G With a PCIe 6.0 SuperNIC and 48-Port Switch](<https://devfeed.tech/articles/cornelis-cn6000-doubles-omni-path-to-800g-with-a-pcie-6-0-supernic-and-48-port-switch-136077.md>)

Original publisher: [Read original article](<https://www.storagereview.com/news/cornelis-cn6000-800g-omni-path-supernic-switch>)

Author: Harold Fritts

Published: 2026-11-16T15:43:20Z

Content type: release

Language: en

Sources: [StorageReview.com](<https://devfeed.tech/sources/storagereview-com.md>)

Topics: [cornelis networks](<https://devfeed.tech/topics/cornelis-networks.md>), [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [pcie](<https://devfeed.tech/topics/pcie.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [gpudirect rdma](<https://devfeed.tech/topics/gpudirect-rdma.md>)

Tags: [400g](<https://devfeed.tech/tags/400g.md>), [accelerators](<https://devfeed.tech/tags/accelerators.md>), [adapter](<https://devfeed.tech/tags/adapter.md>), [adaptive](<https://devfeed.tech/tags/adaptive.md>), [ai](<https://devfeed.tech/tags/ai.md>), [amd](<https://devfeed.tech/tags/amd.md>), [amd-epyc](<https://devfeed.tech/tags/amd-epyc.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [clusters](<https://devfeed.tech/tags/clusters.md>), [cornelis](<https://devfeed.tech/tags/cornelis.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [high-performance](<https://devfeed.tech/tags/high-performance.md>), [high-performance-computing-hpc](<https://devfeed.tech/tags/high-performance-computing-hpc.md>), [hpc](<https://devfeed.tech/tags/hpc.md>), [networking](<https://devfeed.tech/tags/networking.md>), [pcie-6-0](<https://devfeed.tech/tags/pcie-6-0.md>), [rocev2](<https://devfeed.tech/tags/rocev2.md>), [standard](<https://devfeed.tech/tags/standard.md>), [switch](<https://devfeed.tech/tags/switch.md>), [switches](<https://devfeed.tech/tags/switches.md>)

### AI overview

Cornelis's CN6000 brings 800G connectivity to AI and HPC clusters through an Omni-Path SuperNIC, a 48-port Omni-Path switch, and an Ethernet SuperNIC for RoCEv2 networks. The Omni-Path adapter uses PCIe 6.0 x16 with two 400Gb/s ports; its 1.6-billion-message-per-second rate and sub-microsecond latency are design targets. Cornelis reports 0.95µs MPI latency versus 1.2µs for NVIDIA InfiniBand XDR in its internal 8-byte ping-pong test, which the article did not audit; test conditions for the message-rate figure were not published. The switch provides 76.8Tb/s full-duplex capacity and can be configured as two independent 400G planes. Omni-Path supports common HPC software and shares the CN5000 host driver; the Ethernet adapter connects to standard 800G switches using RoCEv2. Ultra Ethernet is in preview and NVMe-oF over RoCEv2 is on the roadmap, with no dates given. Hardware, software, and adapters are available for evaluation and early deployments; OEM systems are expected from Q1 2027.

### Source excerpt

Cornelis has announced the CN6000 family, an 800G networking platform for AI and high-performance computing (HPC) clusters that doubles the bandwidth of the company's CN5000 generation. The family has three parts: an Omni-Path SuperNIC, a 48-port Omni-Path switch, and an Ethernet SuperNIC that runs RoCEv2 against standard 800G switches. Cornelis says the adapters, switch, and The post Cornelis CN6000 Doubles Omni-Path to 800G With a PCIe 6.0 SuperNIC and 48-Port Switch appeared first on StorageReview.com.

## Orthogonal Code-Based Parallel Signaling for PLCA in Shared-Medium Ethernet

DevFeed: [Orthogonal Code-Based Parallel Signaling for PLCA in Shared-Medium Ethernet](<https://devfeed.tech/articles/orthogonal-code-based-parallel-signaling-for-plca-in-shared-medium-ethernet-131580.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.15577>)

Author: Mohamed Elattar, Ahmed Abdelkader, Zhe Lou, Georg Carle

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Networking](<https://devfeed.tech/sources/arxiv-networking.md>)

Topics: [Ethernet](<https://devfeed.tech/topics/ethernet.md>)

Tags: [analysis](<https://devfeed.tech/tags/analysis.md>), [cs-ni](<https://devfeed.tech/tags/cs-ni.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [latency](<https://devfeed.tech/tags/latency.md>), [network](<https://devfeed.tech/tags/network.md>), [node](<https://devfeed.tech/tags/node.md>), [overhead](<https://devfeed.tech/tags/overhead.md>), [parallel](<https://devfeed.tech/tags/parallel.md>), [quantization](<https://devfeed.tech/tags/quantization.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [traffic](<https://devfeed.tech/tags/traffic.md>), [transmission](<https://devfeed.tech/tags/transmission.md>)

### AI overview

OC-PLCA replaces PLCA's sequential transmission-opportunity signaling with simultaneous Walsh-Hadamard code signaling: each node with pending traffic sends its unique code, enabling nodes to identify the active set so the cycle can serve only active nodes and avoid idle opportunities. The work also examines quantization effects and proposes grouping to retain unambiguous detection for arbitrary network sizes. Theoretical analysis reports 88% lower signaling overhead for eight nodes, with larger reductions as network size grows; simulations report lower observed access latency across tested loads and heterogeneous node activity.

### Source excerpt

arXiv:2611.15577v1 Announce Type: new Abstract: Physical Layer Collision Avoidance (PLCA) enables collision-free channel access in shared-medium Ethernet by granting each node an exclusive transmission opportunity (TO) in a sequential, cyclic manner. This sequential, node-based access structure, however, is the root cause of PLCA's fundamental limitations, causing access latency to grow with network size and idle TOs to accumulate when traffic is sparse, a limitation inherited by existing PLCA extensions. To address this, we propose OC-PLCA, an orthogonal code-based parallel signaling mechanism for PLCA. Rather than signaling sequentially, all nodes with pending traffic transmit their unique Walsh-Hadamard code simultaneously to signal their intent to transmit. Since these codes are mutually orthogonal, each node independently identifies the complete set of active nodes, allowing the transmission cycle to include only active nodes and eliminating idle TOs entirely. We further investigate the impact of quantization on active node detection, and propose a grouping-based extension that preserves unambiguous detection for networks of an arbitrary number of nodes. Theoretical analysis shows that OC-PLCA reduces signaling overhead by 88% for an 8-node network, with the reduction growing as network size increases. Simulation results confirm significant reductions in observed access latency across a range of network loads and heterogeneous node activity.

## Renesas RZ/T2N Cortex-A55/R52 industrial MPU supports four-axis motor control, TSN Ethernet, EtherCAT...

DevFeed: [Renesas RZ/T2N Cortex-A55/R52 industrial MPU supports four-axis motor control, TSN Ethernet, EtherCAT...](<https://devfeed.tech/articles/renesas-rz-t2n-cortex-a55-r52-industrial-mpu-supports-four-axis-motor-control-tsn-ethernet-ethercat-128996.md>)

Original publisher: [Read original article](<https://www.cnx-software.com/2026/11/11/renesas-rz-t2n-cortex-a55-r52-industrial-mpu-supports-four-axis-motor-control-tsn-ethernet-ethercat/>)

Author: Debashis Das

Published: 2026-11-11T04:00:29Z

Content type: article

Language: en

Sources: [CNX Software - Embedded Systems News](<https://devfeed.tech/sources/cnx-software-embedded-systems-news.md>)

Topics: [renesas](<https://devfeed.tech/topics/renesas.md>), [cortex a55](<https://devfeed.tech/topics/cortex-a55.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [Modbus](<https://devfeed.tech/topics/modbus.md>)

Tags: [application](<https://devfeed.tech/tags/application.md>), [applications](<https://devfeed.tech/tags/applications.md>), [arm](<https://devfeed.tech/tags/arm.md>), [automation](<https://devfeed.tech/tags/automation.md>), [axis](<https://devfeed.tech/tags/axis.md>), [chip](<https://devfeed.tech/tags/chip.md>), [control](<https://devfeed.tech/tags/control.md>), [cortex](<https://devfeed.tech/tags/cortex.md>), [cortex-a55](<https://devfeed.tech/tags/cortex-a55.md>), [development-board](<https://devfeed.tech/tags/development-board.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [industrial](<https://devfeed.tech/tags/industrial.md>), [linux](<https://devfeed.tech/tags/linux.md>), [motion](<https://devfeed.tech/tags/motion.md>), [motor](<https://devfeed.tech/tags/motor.md>), [motor-control](<https://devfeed.tech/tags/motor-control.md>), [network](<https://devfeed.tech/tags/network.md>), [processors](<https://devfeed.tech/tags/processors.md>), [protocols](<https://devfeed.tech/tags/protocols.md>), [pwm](<https://devfeed.tech/tags/pwm.md>), [real-time](<https://devfeed.tech/tags/real-time.md>), [renesas](<https://devfeed.tech/tags/renesas.md>), [renesas-mcu](<https://devfeed.tech/tags/renesas-mcu.md>), [robotics](<https://devfeed.tech/tags/robotics.md>), [yocto](<https://devfeed.tech/tags/yocto.md>)

### AI overview

The Renesas RZ/T2N combines Linux-capable application processing, real-time control and industrial networking on one MPU. It has up to two 1.2 GHz Cortex-A55 cores and two 800 MHz Cortex-R52 cores, and supports up to four-axis motor control with PWM, encoder interfaces and trigonometric accelerators. Its three-port Gigabit Ethernet switch supports TSN; EtherCAT, PROFINET RT/IRT and EtherNet/IP are listed, while EtherCAT slave control and some redundancy features depend on the variant. The 14-part family varies in core count and feature availability. Software support includes Renesas FSP and e² studio for real-time applications, plus a Linux BSP with kernel 6.12 and Yocto 5.0. Functional safety software is planned for H2 2027.

### Source excerpt

Renesas RZ/T2N industrial MPU supports up to four-axis motor control and includes PWM timers, encoder interfaces, and trigonometric accelerators. It's designed for industrial robots, programmable logic controllers (PLCs), motion controllers, and multi-axis servo drives. The MPU features up to two 1.2 GHz Arm Cortex-A55 cores for application processing and two 800 MHz Cortex-R52 cores for real-time control. It also integrates a three-port Gigabit Ethernet switch with TSN support and industrial Ethernet protocols such as EtherCAT, PROFINET RT/IRT, and EtherNet/IP. This feature set allows real-time motor control, industrial networking, and Linux applications to run on a single chip, reducing the need for separate processors. The MPU can also work as an industrial Ethernet network controller or device, depending on the application. Renesas RZ/T2N specifications: CPU Up to 2x Arm Cortex-A55 (64-bit) cores clocked at 1.2 GHz with 32KB L1 instruction/data cache, 128KB L2 cache per core, and 256KB L3 cache 2x [...] The post Renesas RZ/T2N Cortex-A55/R52 industrial MPU supports four-axis motor control, TSN Ethernet, EtherCAT... appeared first on CNX Software - Embedded Systems News.

## Olimex UEXT-CH390 SPI to Ethernet module targets Raspberry Pi and ESP32 MCU boards running ESPHome

DevFeed: [Olimex UEXT-CH390 SPI to Ethernet module targets Raspberry Pi and ESP32 MCU boards running ESPHome](<https://devfeed.tech/articles/olimex-uext-ch390-spi-to-ethernet-module-targets-raspberry-pi-and-esp32-mcu-boards-running-esphome-120335.md>)

Original publisher: [Read original article](<https://www.cnx-software.com/2026/11/06/olimex-uext-ch390-spi-to-ethernet-module-targets-raspberry-pi-and-esp32-mcu-boards-running-esphome/>)

Author: Debashis Das

Published: 2026-11-06T02:48:49Z

Content type: news

Language: en

Sources: [CNX Software - Embedded Systems News](<https://devfeed.tech/sources/cnx-software-embedded-systems-news.md>)

Topics: [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [esphome](<https://devfeed.tech/topics/esphome.md>), [ESP32](<https://devfeed.tech/topics/esp32.md>), [Embedded Software Dev](<https://devfeed.tech/topics/embedded-software-dev.md>), [Embedded Systems](<https://devfeed.tech/topics/embedded-systems.md>), [Home Assistant](<https://devfeed.tech/topics/home-assistant.md>), [ESP32-C3](<https://devfeed.tech/topics/esp32-c3.md>), [ESP32-C6](<https://devfeed.tech/topics/esp32-c6.md>), [ESP32-S3](<https://devfeed.tech/topics/esp32-s3.md>), [Arduino](<https://devfeed.tech/topics/arduino.md>)

Tags: [arduino](<https://devfeed.tech/tags/arduino.md>), [controller](<https://devfeed.tech/tags/controller.md>), [esp32](<https://devfeed.tech/tags/esp32.md>), [esphome](<https://devfeed.tech/tags/esphome.md>), [espressif](<https://devfeed.tech/tags/espressif.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [home-assistant](<https://devfeed.tech/tags/home-assistant.md>), [iot](<https://devfeed.tech/tags/iot.md>), [iot-sensors](<https://devfeed.tech/tags/iot-sensors.md>), [kicad](<https://devfeed.tech/tags/kicad.md>), [microcontroller](<https://devfeed.tech/tags/microcontroller.md>), [module](<https://devfeed.tech/tags/module.md>), [olimex](<https://devfeed.tech/tags/olimex.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [raspberry-pi](<https://devfeed.tech/tags/raspberry-pi.md>), [spi](<https://devfeed.tech/tags/spi.md>)

### AI overview

Olimex's UEXT-CH390 adds 10/100 Mbps Ethernet to microcontrollers over SPI, using the WCH CH390's integrated MAC and PHY so the host does not need its own Ethernet MAC. Unlike controllers with hardware TCP/IP offloading, the CH390 relies on the host MCU to process network protocols. The 50 x 21 mm module provides RJ45, UEXT and 1.0 mm-pitch pUEXT connectors, runs on 3.3 V, and supports SPI modes 0 and 3. ESPHome support is described for ESP32 devices, and Olimex has added an RP2040 driver tested on a Pico-compatible board with ESPHome 2026.9.0. That setup supports DHCP, Home Assistant discovery, Ethernet OTA updates after initial USB flashing, and a local web interface. Olimex also provides UF2 firmware and a separate Pico SDK example using lwIP.

### Source excerpt

Olimex UEXT-CH390 is a small open-source hardware module that adds 10/100 Mbps wired Ethernet connectivity to microcontroller boards through an SPI interface. It is designed for Raspberry Pi Pico and Pico 2 boards, as well as ESP32-S3, C3, C5, C6, and other microcontrollers without built-in Ethernet connectivity. The module is based on the WCH CH390 Ethernet controller and features an RJ45 connector, a standard UEXT connector, and a smaller 1.0mm-pitch pUEXT connector. It measures 50 x 21 mm and is supported in the Arduino IDE and ESPHome firmware, making it suitable for wired IoT sensors, Home Assistant devices, and other applications where WiFi is unreliable or unwanted. Olimex UEXT-CH390 specifications: Ethernet controller - WCH CH390 with integrated 10/100 Mbps Ethernet MAC and PHY 10BASE-T and 100BASE-TX Ethernet with auto-negotiation Auto-MDIX support for automatic cable detection Built-in 16KB SRAM for packet buffering Factory-programmed MAC address, eliminating the need for an external [...] The post Olimex UEXT-CH390 SPI to Ethernet module targets Raspberry Pi and ESP32 MCU boards running ESPHome appeared first on CNX Software - Embedded Systems News.

## New open source hardware UEXT module adds 100MB Ethernet to PICO, PICO2 and all ESP32s lacking wired Ethernet

DevFeed: [New open source hardware UEXT module adds 100MB Ethernet to PICO, PICO2 and all ESP32s lacking wired Ethernet](<https://devfeed.tech/articles/new-open-source-hardware-uext-module-adds-100mb-ethernet-to-pico-pico2-and-all-esp32s-lacking-wired-ethernet-117115.md>)

Original publisher: [Read original article](<https://blog.adafruit.com/2026/11/04/new-open-source-hardware-uext-module-adds-100mb-ethernet-to-pico-pico2-and-all-esp32s-lacking-wired-ethernet/>)

Author: Anne Barela

Published: 2026-11-04T16:44:08Z

Content type: news

Language: en

Sources: [Adafruit Industries](<https://devfeed.tech/sources/adafruit-industries-makers-hackers-artists-designers-and-engineers.md>)

Topics: [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [kicad](<https://devfeed.tech/topics/kicad.md>), [esphome](<https://devfeed.tech/topics/esphome.md>), [Home Assistant](<https://devfeed.tech/topics/home-assistant.md>)

Tags: [arduino](<https://devfeed.tech/tags/arduino.md>), [bluetooth](<https://devfeed.tech/tags/bluetooth.md>), [controller](<https://devfeed.tech/tags/controller.md>), [driver](<https://devfeed.tech/tags/driver.md>), [esp32](<https://devfeed.tech/tags/esp32.md>), [esp32-s3](<https://devfeed.tech/tags/esp32-s3.md>), [esphome](<https://devfeed.tech/tags/esphome.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [example](<https://devfeed.tech/tags/example.md>), [firmware](<https://devfeed.tech/tags/firmware.md>), [github](<https://devfeed.tech/tags/github.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [kicad](<https://devfeed.tech/tags/kicad.md>), [module](<https://devfeed.tech/tags/module.md>), [open](<https://devfeed.tech/tags/open.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [open-source-hardware](<https://devfeed.tech/tags/open-source-hardware.md>), [pico](<https://devfeed.tech/tags/pico.md>), [raspberry-pi](<https://devfeed.tech/tags/raspberry-pi.md>), [spi](<https://devfeed.tech/tags/spi.md>), [uext](<https://devfeed.tech/tags/uext.md>), [wch](<https://devfeed.tech/tags/wch.md>), [wired](<https://devfeed.tech/tags/wired.md>)

### AI overview

Olimex's UEXT-CH390 adds 100 Mbit wired Ethernet to boards with UEXT or pUEXT connectors, using a WCH CH390 SPI-to-Ethernet controller. It supports Raspberry Pi PICO/PICO2 and ESP32 variants without wired Ethernet MACs, including S3, C3, C5 and C6. Arduino and ESPHome support CH390, but ESPHome's support was limited to ESP32, so Olimex wrote an RP2040 driver. The excerpt describes an ESPHome example pinned to a version of the Ethernet component and tested with ESPHome 2026.9.0, a flashable UF2 that appears in Home Assistant, and a Pico SDK/lwIP web-server example. KiCad sources, schematic and firmware are published. The 50 x 21 mm module has both UEXT and smaller pUEXT headers; it is listed at 4.95 EUR for pre-order, with shipping from 9 November 2026.

### Source excerpt

Olimex has announced UEXT-CH390 on the company blog, a small open source hardware module that adds 100 Mbit wired Ethernet to any board with a UEXT or pUEXT connector. UEXT is Olimex's own connector standard for add-on boards. The module uses the WCH CH390 SPI-to-Ethernet controller, measures 50 x 21 mm, and carries both the [...]

## $21 ESP32-P4X and ESP32-C61 full-featured board offers Ethernet, WiFi 6, RS485 CAN Bus, MIPI CSI/DSI, and more

DevFeed: [$21 ESP32-P4X and ESP32-C61 full-featured board offers Ethernet, WiFi 6, RS485 CAN Bus, MIPI CSI/DSI, and more](<https://devfeed.tech/articles/21-esp32-p4x-and-esp32-c61-full-featured-board-offers-ethernet-wifi-6-rs485-can-bus-mipi-csi-dsi-and-more-116382.md>)

Original publisher: [Read original article](<https://www.cnx-software.com/2026/11/04/21-esp32-p4x-and-esp32-c61-full-featured-board-offers-ethernet-wifi-6-rs485-can-bus-mipi-csi-dsi-and-more/>)

Author: Jean-Luc Aufranc (CNXSoft)

Published: 2026-11-04T10:05:31Z

Content type: news

Language: en

Sources: [CNX Software - Embedded Systems News](<https://devfeed.tech/sources/cnx-software-embedded-systems-news.md>)

Topics: [ESP32-P4](<https://devfeed.tech/topics/esp32-p4.md>), [PSRAM](<https://devfeed.tech/topics/psram.md>), [RISC-V](<https://devfeed.tech/topics/riscv.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [USB](<https://devfeed.tech/topics/usb.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ble](<https://devfeed.tech/tags/ble.md>), [board](<https://devfeed.tech/tags/board.md>), [bus](<https://devfeed.tech/tags/bus.md>), [can-bus](<https://devfeed.tech/tags/can-bus.md>), [csi](<https://devfeed.tech/tags/csi.md>), [development-board](<https://devfeed.tech/tags/development-board.md>), [display](<https://devfeed.tech/tags/display.md>), [esp32](<https://devfeed.tech/tags/esp32.md>), [esp32-p4](<https://devfeed.tech/tags/esp32-p4.md>), [espressif](<https://devfeed.tech/tags/espressif.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [fast](<https://devfeed.tech/tags/fast.md>), [gateways](<https://devfeed.tech/tags/gateways.md>), [h-264](<https://devfeed.tech/tags/h-264.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [hp](<https://devfeed.tech/tags/hp.md>), [industrial](<https://devfeed.tech/tags/industrial.md>), [jpeg](<https://devfeed.tech/tags/jpeg.md>), [mipi](<https://devfeed.tech/tags/mipi.md>), [mipi-csi](<https://devfeed.tech/tags/mipi-csi.md>), [rs485](<https://devfeed.tech/tags/rs485.md>), [smart-speaker](<https://devfeed.tech/tags/smart-speaker.md>), [touchscreen](<https://devfeed.tech/tags/touchscreen.md>), [video](<https://devfeed.tech/tags/video.md>), [wifi](<https://devfeed.tech/tags/wifi.md>), [wifi-6](<https://devfeed.tech/tags/wifi-6.md>), [wireless](<https://devfeed.tech/tags/wireless.md>)

### AI overview

The Wireless Tag WT99P461-S1 is a $21 development board built around the WT01P461-S1 module, with ESP32-P4X (confirmed by Wireless Tag as ESP32-P4 v3.2) and ESP32-C61. Compared with the ESP32P4C61-TINY, it adds 10/100 Mbps Ethernet, RS-485, CAN, an onboard microphone and speaker connector; it also has Wi-Fi 6/Bluetooth, microSD, MIPI DSI and CSI, and USB Type-A and Type-C ports. It has 32 MB PSRAM and supports a 7-inch touchscreen. Preloaded ESP-Brookesia firmware includes AI vision demos, MP3 and MJPEG playback, camera support, and network connectivity; H.264 playback is not yet supported. The board costs $21, or $42.50 with a 7-inch display kit, before shipping and taxes. The product page is unclear about the processor version, and the article reports that useful project resources were difficult to find.

### Source excerpt

Wireless Tag WT99P461-S1 is an ESP32-P4X and ESP32-C61 board based on the same WT01P461-S1 module as found on the ESP32P4C61-TINY board introduced last June. The main differences are that the WT99P461-S1 is a full-featured development board, adding features such as Fast Ethernet, RS-485, CAN bus, an on-board microphone, and a speaker connector. It still comes with a microSD card slot, MIPI DSI/CSI connectors, and other features, making it suitable for industrial HMIs, gateways, and control applications. WT99P461-S1 specifications: Wireless Module - Wireless-Tag WT01P461-S1 (25 x 25mm) Microcontroller - ESP32-P4X (ESP32-P4 v3.2) MCU Dual-core RISC-V microcontroller @ 360/400 MHz with AI instructions extension and single-precision FPU Single-RISC-V LP (Low-power) MCU core @ up to 40 MHz GPU - 2D Pixel Processing Accelerator (PPA) VPU - H.264 and JPEG codec support Memory - 768 KB HP L2MEM, 32 KB LP SRAM, 8 KB TCM, 32MB PSRAM Storage - 128 KB HP ROM, 16 [...] The post $21 ESP32-P4X and ESP32-C61 full-featured board offers Ethernet, WiFi 6, RS485 CAN Bus, MIPI CSI/DSI, and more appeared first on CNX Software - Embedded Systems News.

## Progressive-Observation Physical-Layer Fingerprinting for Efficient Open-Set Ethernet Authentication

DevFeed: [Progressive-Observation Physical-Layer Fingerprinting for Efficient Open-Set Ethernet Authentication](<https://devfeed.tech/articles/progressive-observation-physical-layer-fingerprinting-for-efficient-open-set-ethernet-authentication-107613.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.38074>)

Author: Chenming Zhang, Jiaxiang Geng, Aiqun Hu, Azzedine Boukerche, Peng Sun

Published: 2026-10-30T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>), [arXiv: Networking](<https://devfeed.tech/sources/arxiv-networking.md>)

Topics: [Access Control](<https://devfeed.tech/topics/access-control.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [smartdevices](<https://devfeed.tech/topics/smartdevices.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [acceptance](<https://devfeed.tech/tags/acceptance.md>), [adaptive](<https://devfeed.tech/tags/adaptive.md>), [algorithm](<https://devfeed.tech/tags/algorithm.md>), [authentication](<https://devfeed.tech/tags/authentication.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [cnn](<https://devfeed.tech/tags/cnn.md>), [controller](<https://devfeed.tech/tags/controller.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ni](<https://devfeed.tech/tags/cs-ni.md>), [data](<https://devfeed.tech/tags/data.md>), [devices](<https://devfeed.tech/tags/devices.md>), [edge](<https://devfeed.tech/tags/edge.md>), [eess-sp](<https://devfeed.tech/tags/eess-sp.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [features](<https://devfeed.tech/tags/features.md>), [fixed](<https://devfeed.tech/tags/fixed.md>), [full](<https://devfeed.tech/tags/full.md>), [internet-of-things](<https://devfeed.tech/tags/internet-of-things.md>), [layer](<https://devfeed.tech/tags/layer.md>), [lightweight](<https://devfeed.tech/tags/lightweight.md>), [matching](<https://devfeed.tech/tags/matching.md>), [open](<https://devfeed.tech/tags/open.md>)

### AI overview

The method reduces Ethernet device authentication cost by adapting waveform length to each query. A lightweight CNN rejects clearly unknown devices from a short observation; spectral matching accepts clearly registered devices from a slightly longer one; ambiguous queries use the full waveform. To handle day-to-day fingerprint drift without retraining the CNN, calibration realigns features and thresholds using a few current-day references from registered devices. In tests on 100BASE-TX devices with two controller types on held-out days, false rejection and acceptance rates were no higher than with fixed-length processing. Average waveform data per query fell 75.1%, and CPU time on a Raspberry Pi 5 fell 73.7%; at the same monitoring bandwidth, the authors estimate an edge node can handle about four times as many queries.

### Source excerpt

arXiv:2610.38074v1 Announce Type: cross Abstract: Physical-layer fingerprinting (PLF) has emerged as a promising authentication approach for Internet of Things devices. However, most PLF systems acquire and process a fixed-length waveform for every authentication query, burdening resource-constrained edge nodes. To address this issue, we propose a progressive-observation PLF method that adapts the waveform length to each query. Specifically, a lightweight convolutional neural network (CNN) first rejects clearly unknown devices from a short observation. Spectral matching then accepts clearly registered devices from a slightly longer observation. Only ambiguous queries proceed to the same matching on the full-length waveform, reducing both data acquisition and computation. Because fingerprints drift across days, we also design an adaptive calibration algorithm to avoid retraining the CNN. The algorithm realigns the features and decision thresholds using a few current-day references from registered devices. We evaluate the method with 100BASE-TX Ethernet devices of two controller types on held-out days. Compared with fixed-length processing, it keeps the false rejection and false acceptance rates equal or lower for both controller types on both days. It also reduces the waveform data per query by 75.1% and the CPU processing time on a Raspberry Pi 5 by 73.7% on average. With the same monitoring bandwidth, an edge node can therefore serve about four times as many authentication queries.

## MCCL: Training Pytorch models across a fleet of Macs at Skywalker Sound

DevFeed: [MCCL: Training Pytorch models across a fleet of Macs at Skywalker Sound](<https://devfeed.tech/articles/mccl-training-pytorch-models-across-a-fleet-of-macs-at-skywalker-sound-107579.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.39029>)

Author: Kevin Sasso, Chris Fiegel, Francis Aitken, Nicolas Tsingos, Nicholas Tong

Published: 2026-10-30T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Distributed Computing](<https://devfeed.tech/sources/arxiv-distributed-computing.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [distributed-training](<https://devfeed.tech/topics/distributed-training.md>), [Distributed Systems & Parallel Computing](<https://devfeed.tech/topics/distributed-systems-parallel-computing.md>), [PyTorch](<https://devfeed.tech/topics/pytorch.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>)

Tags: [cloud-compute](<https://devfeed.tech/tags/cloud-compute.md>), [compute-costs](<https://devfeed.tech/tags/compute-costs.md>), [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [distributed-training](<https://devfeed.tech/tags/distributed-training.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [gpu-clusters](<https://devfeed.tech/tags/gpu-clusters.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [mac](<https://devfeed.tech/tags/mac.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [matrix](<https://devfeed.tech/tags/matrix.md>), [metal](<https://devfeed.tech/tags/metal.md>), [mps](<https://devfeed.tech/tags/mps.md>), [node](<https://devfeed.tech/tags/node.md>), [nvidia-a100](<https://devfeed.tech/tags/nvidia-a100.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [performance](<https://devfeed.tech/tags/performance.md>), [pytorch](<https://devfeed.tech/tags/pytorch.md>), [scalability](<https://devfeed.tech/tags/scalability.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

MCCL adds an open-source Metal Performance Shaders backend for multi-node PyTorch training over switched Ethernet, addressing PyTorch's lack of an accelerated MPS distributed backend. In tests on up to 24 Macs, large all-reduce operations achieved 98-99.7% of TCP goodput, and a 113.5-million-parameter Vision Transformer scaled from 198 to 4,568 samples per second at 95.9% parallel efficiency. Against an NVIDIA A100, one M4 Max reportedly delivered about 1.7 times its FP32 matrix-multiplication throughput; 5.5 Macs matched the A100 using TF32. The abstract estimates that using 100 idle Macs for 12 hours nightly could offset about $200,000 in annual equivalent cloud compute costs. These are reported benchmarks and estimates; the supplied text is an abstract, not full paper evidence.

### Source excerpt

arXiv:2610.39029v1 Announce Type: new Abstract: Skywalker Sound, a premier audio post-production facility, operates more than 200 M4 Max Mac Studios on a 10Gb Ethernet network. When not in use by artists, these machines sit idle, representing massive untapped unified memory and compute capacity. While these fleets are excellent candidates for distributed machine learning alongside existing NVIDIA GPU clusters, PyTorch lacks an accelerated distributed backend for Apple's Metal Performance Shaders (MPS). To bridge this gap, we introduce the Metal Communication Collective Library (MCCL), the first open-source MPS backend for multi-node PyTorch training over switched Ethernet. We evaluate its performance and scalability across 24 nodes, demonstrating that large all-reduce operations reach 98-99.7% of TCP goodput. Training a 113.5M-parameter Vision Transformer (ViT) scales from 198 to 4,568 samples/s, achieving 95.9% parallel efficiency. Benchmarked against an NVIDIA A100 GPU, one M4 Max Mac equals roughly 1.7x the throughput of FP32 matrix multiplication, and 5.5 Macs match an A100 using TF32. Harnessing 100 idle Macs for 12 hours nightly offsets approximately $200,000 annually in equivalent cloud compute costs. Our results demonstrate that existing Mac infrastructure can be seamlessly repurposed for high performance distributed training without abandoning the standard PyTorch codebase.

## The EVPN trust challenge: How to detect and contain internal MAC hijacks

DevFeed: [The EVPN trust challenge: How to detect and contain internal MAC hijacks](<https://devfeed.tech/articles/the-evpn-trust-challenge-how-to-detect-and-contain-internal-mac-hijacks-103528.md>)

Original publisher: [Read original article](<https://blog.apnic.net/2026/10/28/the-evpn-trust-challenge-how-to-detect-and-contain-internal-mac-hijacks/>)

Author: Sabina Anja

Published: 2026-10-27T22:32:41Z

Content type: article

Language: en

Sources: [APNIC Blog](<https://devfeed.tech/sources/apnic-blog.md>)

Topics: [evpn](<https://devfeed.tech/topics/evpn.md>), [Routing Security](<https://devfeed.tech/topics/routing-security.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [rate-limiting](<https://devfeed.tech/topics/rate-limiting.md>), [observability](<https://devfeed.tech/topics/observability.md>), [Monitoring](<https://devfeed.tech/topics/monitoring.md>), [trust](<https://devfeed.tech/topics/trust.md>)

Tags: [arp](<https://devfeed.tech/tags/arp.md>), [attack](<https://devfeed.tech/tags/attack.md>), [bgp](<https://devfeed.tech/tags/bgp.md>), [denial-of-service](<https://devfeed.tech/tags/denial-of-service.md>), [detection](<https://devfeed.tech/tags/detection.md>), [dhcp](<https://devfeed.tech/tags/dhcp.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [evpn](<https://devfeed.tech/tags/evpn.md>), [fabric](<https://devfeed.tech/tags/fabric.md>), [guest-post](<https://devfeed.tech/tags/guest-post.md>), [mac](<https://devfeed.tech/tags/mac.md>), [mac-address](<https://devfeed.tech/tags/mac-address.md>), [mobility](<https://devfeed.tech/tags/mobility.md>), [monitoring](<https://devfeed.tech/tags/monitoring.md>), [network-operators](<https://devfeed.tech/tags/network-operators.md>), [route](<https://devfeed.tech/tags/route.md>), [security](<https://devfeed.tech/tags/security.md>), [spoofing](<https://devfeed.tech/tags/spoofing.md>), [tech-matters](<https://devfeed.tech/tags/tech-matters.md>), [traffic](<https://devfeed.tech/tags/traffic.md>), [trust](<https://devfeed.tech/tags/trust.md>), [visibility](<https://devfeed.tech/tags/visibility.md>), [vpn](<https://devfeed.tech/tags/vpn.md>)

### AI overview

EVPN MAC mobility can turn a forged source MAC into a fabric-wide forwarding change: a leaf advertises the learned address with a higher sequence number, and other PEs prefer that route until the legitimate host advertises again. The article describes interception, denial of service, control-plane churn and possible IP hijacking, and notes that duplicate-MAC safeguards vary by platform: Juniper suppresses until manually cleared, while Nokia SR OS flushes and retries after 90 minutes. Detection recommendations include exporting route-reflector updates via BMP, tracking MAC moves over longer periods, and investigating sequence jumps, unexpected sticky flags, tie-break flapping and routes that conflict with access-layer records. It emphasizes validating host identity at the edge with controls such as MAC limits, DHCP snooping, ARP/ND inspection, IEEE 802.1X and sticky entries for critical systems.

### Source excerpt

Guest Post: EVPN MAC mobility enables seamless workload movement, but it can also amplify the impact of MAC spoofing. A forged MAC address can redirect traffic across the fabric through BGP, making access-layer controls, route-reflector visibility, and MAC-move monitoring critical defences.

## 11 Networking Concepts Explained in 11 Minutes

DevFeed: [11 Networking Concepts Explained in 11 Minutes](<https://devfeed.tech/articles/11-networking-concepts-explained-in-11-minutes-97839.md>)

Original publisher: [Read original article](<https://newsletter.systemdesign.one/p/computer-networking-fundamentals-for-software-engineers>)

Author: Neo Kim

Published: 2026-10-23T16:06:14Z

Content type: article

Language: en

Sources: [System Design Newsletter](<https://devfeed.tech/sources/system-design-newsletter.md>)

Topics: [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [network troubleshooting commands](<https://devfeed.tech/topics/network-troubleshooting-commands.md>), [Cloud Networking](<https://devfeed.tech/topics/cloud-networking.md>), [CIDR and subnets](<https://devfeed.tech/topics/cidr-and-subnets.md>), [MTU mismatch](<https://devfeed.tech/topics/mtu-mismatch.md>), [Tool](<https://devfeed.tech/topics/tool.md>)

Tags: [concepts](<https://devfeed.tech/tags/concepts.md>), [networking](<https://devfeed.tech/tags/networking.md>)

### AI overview

The excerpt explains how local and routed network communication works and how to investigate common connectivity issues. Ethernet carries IP packets in frames; switches learn MAC-to-port mappings and flood frames when a destination is unknown, while routers forward IP packets between networks. DHCP assigns leased addresses and settings such as the gateway and DNS servers; static addresses provide predictable assignments. Devices send remote traffic to a default gateway, and routers use static routes or OSPF to direct packets. Static routes require manual updates when paths change, while OSPF shares topology information and recalculates paths after failures, with convergence depending on detection, topology, and configuration. Wi-Fi performance varies with band, range, interference, signal quality, and retransmissions. ICMP reports errors and supports diagnostics but can be blocked or dropped and does not repair connections. Ping measures ICMP response time and missing replies; a successful ping does not establish that an application service is available. The excerpt recommends tools including `ip route`, `mtr`, `tcpdump`, and Wireshark for examining network behavior.

### Source excerpt

#195: Part 2 - Break Into Computer Networking

## Qotom Q51551UT 4x 10GbE and 6x 2.5GbE Mini PC Review

DevFeed: [Qotom Q51551UT 4x 10GbE and 6x 2.5GbE Mini PC Review](<https://devfeed.tech/articles/qotom-q51551ut-4x-10gbe-and-6x-2-5gbe-mini-pc-review-96141.md>)

Original publisher: [Read original article](<https://www.servethehome.com/qotom-q51551ut-4x-10gbe-and-6x-2-5gbe-intel-mini-pc-review/>)

Author: Ryan Smith

Published: 2026-10-22T17:52:06Z

Content type: article

Language: en

Sources: [ServeTheHome](<https://devfeed.tech/sources/servethehome.md>)

Topics: [10 Gigabit Ethernet](<https://devfeed.tech/topics/10gbe-networking.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>)

Tags: [10gbe](<https://devfeed.tech/tags/10gbe.md>), [2-5gbe](<https://devfeed.tech/tags/2-5gbe.md>), [home-office-entry](<https://devfeed.tech/tags/home-office-entry.md>), [meteor-lake](<https://devfeed.tech/tags/meteor-lake.md>), [mini-pc](<https://devfeed.tech/tags/mini-pc.md>), [networking](<https://devfeed.tech/tags/networking.md>), [qotom](<https://devfeed.tech/tags/qotom.md>), [review](<https://devfeed.tech/tags/review.md>), [workstation](<https://devfeed.tech/tags/workstation.md>)

### AI overview

The Qotom Q51551UT combines ten Ethernet ports with a Core Ultra 5 125H and display outputs, making it suitable for network focused workloads such as a software firewall or edge computer. Six Intel I226-V ports support up to 2.5GbE, while two SFP+ ports use an Intel X710-BM2 and can take 1GbE or 10GbE modules. Two copper 10GbE ports use the older Intel X540-AT2; they do not support 2.5Gbps or 5Gbps intermediate rates. Qotom also specifies bandwidth sharing that prevents all NICs from reaching full speed simultaneously. The system has HDMI 2.0 and DisplayPort 1.4 outputs driven by the integrated GPU, plus active blower cooling. This excerpt covers the external hardware and notes that the review will discuss bandwidth further; it does not provide performance results.

### Source excerpt

Today we are taking a look at a network-focused industrial mini-PC from Qotom, the Q51551UT. The small box goes big on networking, with 10 Ethernet ports that support data rates as high as 10Gbps The post Qotom Q51551UT 4x 10GbE and 6x 2.5GbE Mini PC Review appeared first on ServeTheHome.

## GL.iNet Mango 2 mini travel router features AC1300 WiFi 5 and dual Gigabit Ethernet

DevFeed: [GL.iNet Mango 2 mini travel router features AC1300 WiFi 5 and dual Gigabit Ethernet](<https://devfeed.tech/articles/gl-inet-mango-2-mini-travel-router-features-ac1300-wifi-5-and-dual-gigabit-ethernet-95251.md>)

Original publisher: [Read original article](<https://www.cnx-software.com/2026/10/22/gl-inet-mango-2-mini-travel-router-features-ac1300-wifi-5-and-dual-gigabit-ethernet/>)

Author: Debashis Das

Published: 2026-10-22T09:00:05Z

Content type: news

Language: en

Sources: [CNX Software - Embedded Systems News](<https://devfeed.tech/sources/cnx-software-embedded-systems-news.md>)

Topics: [GL.iNet Mango 2](<https://devfeed.tech/topics/gl-inet-mango-2.md>), [GL-MG1300](<https://devfeed.tech/topics/gl-mg1300.md>), [portable router](<https://devfeed.tech/topics/portable-router.md>), [GL.iNet](<https://devfeed.tech/topics/gl-inet.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [USB](<https://devfeed.tech/topics/usb.md>)

Tags: [gigabit-ethernet](<https://devfeed.tech/tags/gigabit-ethernet.md>), [gl-inet](<https://devfeed.tech/tags/gl-inet.md>), [gl-inet-mango](<https://devfeed.tech/tags/gl-inet-mango.md>), [gl-inet-mango-2](<https://devfeed.tech/tags/gl-inet-mango-2.md>), [gl-mg1300](<https://devfeed.tech/tags/gl-mg1300.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [linux](<https://devfeed.tech/tags/linux.md>), [mango-2](<https://devfeed.tech/tags/mango-2.md>), [mediatek](<https://devfeed.tech/tags/mediatek.md>), [mediatek-wi-fi](<https://devfeed.tech/tags/mediatek-wi-fi.md>), [mini](<https://devfeed.tech/tags/mini.md>), [mini-travel-router](<https://devfeed.tech/tags/mini-travel-router.md>), [networking](<https://devfeed.tech/tags/networking.md>), [openvpn](<https://devfeed.tech/tags/openvpn.md>), [openwrt](<https://devfeed.tech/tags/openwrt.md>), [processor](<https://devfeed.tech/tags/processor.md>), [router](<https://devfeed.tech/tags/router.md>), [travel](<https://devfeed.tech/tags/travel.md>), [usb](<https://devfeed.tech/tags/usb.md>), [usb-c](<https://devfeed.tech/tags/usb-c.md>), [video](<https://devfeed.tech/tags/video.md>), [vpn](<https://devfeed.tech/tags/vpn.md>), [wifi](<https://devfeed.tech/tags/wifi.md>), [wifi-5](<https://devfeed.tech/tags/wifi-5.md>), [wired](<https://devfeed.tech/tags/wired.md>), [wireguard](<https://devfeed.tech/tags/wireguard.md>)

### AI overview

The GL.iNet Mango 2 (GL-MG1300) is a 100-gram travel router for sharing hotel WiFi, wired Ethernet, or mobile internet with multiple devices. It offers WiFi 5 up to 400 Mbps at 2.4 GHz and 866 Mbps at 5 GHz, dual Gigabit Ethernet, USB 3.0, and captive-portal sign-in sharing. VPN throughput is listed at up to 184 Mbps with WireGuard and 95 Mbps with OpenVPN DCO; the router supports client/server modes, Tailscale, multi-tunnel setups, and device- or website-based VPN policies. Cellular access requires a compatible USB modem or smartphone tethering. Compared with the cited Beryl AX and Slate 7, it is a compact entry-level WiFi 5 option with lower wireless speeds. The text lists it at $49.99 through GL.iNet's official AliExpress store and says the company does not sell it in the US due to FCC-related restrictions.

### Source excerpt

GL.iNet Mango 2 (GL-MG1300) mini dual-band WiFi 5 travel router is built around the MediaTek MT7621A dual-core processor. It is designed to share hotel WiFi, wired Ethernet, or mobile internet connections with multiple devices. The router measures just 89 x 63 x 15mm and weighs 100 grams. It supports WiFi speeds of up to 400 Mbps on the 2.4 GHz band and 866 Mbps on the 5 GHz band, along with WireGuard VPN speeds of up to 184 Mbps and OpenVPN speeds of up to 95 Mbps with DCO. Other features include dual Gigabit Ethernet ports, USB 3.0, USB-C power, and others. GL.iNet Mango 2 (GL-MG1300) specifications: Processor - Mediatek MT7621A dual core MIPS 1004Kc processor @ 880Mhz Memory - 128MB DDR3L RAM Storage - 128MB SPI NAND flash + 2MB SPI NOR flash Networking 2x Gigabit Ethernet ports (WAN/LAN) Dual-band 802.11a/b/g/n/ac WiFi 5 2.4 GHz - Up to 400 [...] The post GL.iNet Mango 2 mini travel router features AC1300 WiFi 5 and dual Gigabit Ethernet appeared first on CNX Software - Embedded Systems News.

## Inside the Secretive AWS Networking Lab

DevFeed: [Inside the Secretive AWS Networking Lab](<https://devfeed.tech/articles/inside-the-secretive-aws-networking-lab-87114.md>)

Original publisher: [Read original article](<https://www.servethehome.com/inside-the-secretive-aws-networking-lab/>)

Author: Patrick Kennedy

Published: 2026-10-14T17:15:44Z

Content type: article

Language: en

Sources: [ServeTheHome](<https://devfeed.tech/sources/servethehome.md>)

Topics: [Cryptography](<https://devfeed.tech/topics/cryptography.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [Amazon VPC](<https://devfeed.tech/topics/amazon-vpc.md>)

Tags: [1-6t](<https://devfeed.tech/tags/1-6t.md>), [1-6tbe](<https://devfeed.tech/tags/1-6tbe.md>), [102-4t](<https://devfeed.tech/tags/102-4t.md>), [400g](<https://devfeed.tech/tags/400g.md>), [51-2t](<https://devfeed.tech/tags/51-2t.md>), [ai](<https://devfeed.tech/tags/ai.md>), [aws](<https://devfeed.tech/tags/aws.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [dwdm](<https://devfeed.tech/tags/dwdm.md>), [encryption](<https://devfeed.tech/tags/encryption.md>), [networking](<https://devfeed.tech/tags/networking.md>), [osfp](<https://devfeed.tech/tags/osfp.md>), [qsfp112](<https://devfeed.tech/tags/qsfp112.md>)

### AI overview

A tour of AWS's networking lab covers its in-house switches from 3.2 Tbps to 102.4 Tbps, including a 12.8 Tbps model that encrypts traffic on every port. The article also describes a separate timing network that AWS says keeps time within 20 microseconds, plus management switches and related equipment.

### Source excerpt

We tour a secretive AWS networking lab to see innovation from management switches up to 102.4T switches, encryption, timing, optics, and more The post Inside the Secretive AWS Networking Lab appeared first on ServeTheHome.

## Cisco, Nvidia push ultra-dense networking for Ethernet AI fabrics

DevFeed: [Cisco, Nvidia push ultra-dense networking for Ethernet AI fabrics](<https://devfeed.tech/articles/cisco-nvidia-push-ultra-dense-networking-for-ethernet-ai-fabrics-85942.md>)

Original publisher: [Read original article](<https://www.networkworld.com/article/4233992/cisco-nvidia-push-ultra-dense-networking-for-ethernet-ai-fabrics.html>)

Author: Michael Cooney

Published: 2026-10-13T13:15:28Z

Content type: news

Language: en

Sources: [Network World](<https://devfeed.tech/sources/network-world.md>)

Topics: [AI Networking](<https://devfeed.tech/topics/ai-networking.md>), [Nvidia](<https://devfeed.tech/topics/nvidia.md>), [Spectrum Ethernet](<https://devfeed.tech/topics/spectrum-ethernet.md>), [Cisco](<https://devfeed.tech/topics/cisco.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [NVLink](<https://devfeed.tech/topics/nvlink.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [ai-systems](<https://devfeed.tech/tags/ai-systems.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [artificial-intelligence-network-switches-networking-networking-devices](<https://devfeed.tech/tags/artificial-intelligence-network-switches-networking-networking-devices.md>), [cisco](<https://devfeed.tech/tags/cisco.md>), [edition-us](<https://devfeed.tech/tags/edition-us.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [language-en](<https://devfeed.tech/tags/language-en.md>), [network-switches](<https://devfeed.tech/tags/network-switches.md>), [networking](<https://devfeed.tech/tags/networking.md>), [networking-devices](<https://devfeed.tech/tags/networking-devices.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [ocp](<https://devfeed.tech/tags/ocp.md>), [optical-systems](<https://devfeed.tech/tags/optical-systems.md>), [pluggable-optics](<https://devfeed.tech/tags/pluggable-optics.md>), [scale](<https://devfeed.tech/tags/scale.md>), [storytype-news](<https://devfeed.tech/tags/storytype-news.md>), [summit](<https://devfeed.tech/tags/summit.md>)

### AI overview

Cisco and Nvidia introduced the Ultra-Dense Small Form Factor Pluggable (USFP), an optical module designed for Ethernet AI fabrics. It supports 400 Gb/s-per-lane signals, with a stated path from 6.4T aggregate capacity across 16 lanes to 12.8T across 32 lanes. The companies also announced an expansion of Cisco's Silicon One switching portfolio, including the G302 chip and two fixed-form switches.

### Source excerpt

Cisco and Nvidia have teamed up again. This time it's to deliver a high-capacity pluggable optical system that promises to let customers more quickly and easily build Ethernet AI fabrics. Introduced at the Open Compute Project (OCP) Global Summit 2026 this week, the Ultra-Dense Small Form Factor Pluggable (USFP) module supports 400 Gb/s-per-lane electrical signals, allowing up to 12.8T of capacity for building an infrastructure that spans networking, security, AI systems, and high-performance AI fabrics. The idea is to help customers address the pressure AI infrastructure is placing on the connections between GPUs, switches, security appliances, and other systems, according to Cisco. Increasing bandwidth density can help customers build higher-capacity systems without proportionally increasing the space consumed by connectivity hardware, according to Bill Gartner, senior vice president and general manager of Cisco's optical systems and optics business, and Will Eatherton, senior vice president, networking engineering. "USFP will provide a path supporting both 200G and 400G SerDes based systems and ASICs, helping customers scale from today's 1.6T fabrics," Gartner and Eatherton wrote in a blog post. "With 16 lanes capable of operating at 400 Gb/s per lane, USFP will deliver 6.4T of aggregate capacity, four times the bandwidth density of today's 1.6T pluggables," they added. "The architecture can scale to 32 lanes, providing a path to 12.8T of aggregate capacity. It also creates an opportunity for approximately 1.3x greater power efficiency through the use of 400G SerDes technology, helping operators manage growing power and cooling demands," Gartner and Eatherton wrote. "Our work has been guided by a shared goal: helping customers accelerate AI adoption while maintaining flexibility, openness, and choice," Gartner and Eatherton wrote, referring to Cisco and Nvidia's ongoing collaboration. USFP aims to preserve the serviceability of pluggable optics; modules can be re

## New USFP Pluggable Optics Form Factor for Higher Speeds from Cisco and NVIDIA

DevFeed: [New USFP Pluggable Optics Form Factor for Higher Speeds from Cisco and NVIDIA](<https://devfeed.tech/articles/new-usfp-pluggable-optics-form-factor-for-higher-speeds-from-cisco-and-nvidia-86222.md>)

Original publisher: [Read original article](<https://www.servethehome.com/cisco-ocp-ai-networking-announcements-nvidia/>)

Author: Patrick Kennedy

Published: 2026-10-13T13:00:40Z

Content type: news

Language: en

Sources: [ServeTheHome](<https://devfeed.tech/sources/servethehome.md>)

Topics: [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [Cisco](<https://devfeed.tech/topics/cisco.md>), [Nvidia](<https://devfeed.tech/topics/nvidia.md>), [Nexus Dashboard](<https://devfeed.tech/topics/nexus-dashboard.md>)

Tags: [400g](<https://devfeed.tech/tags/400g.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [ai-networking](<https://devfeed.tech/tags/ai-networking.md>), [cisco](<https://devfeed.tech/tags/cisco.md>), [custom-silicon](<https://devfeed.tech/tags/custom-silicon.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [hyperscale](<https://devfeed.tech/tags/hyperscale.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [marvell](<https://devfeed.tech/tags/marvell.md>), [n9000](<https://devfeed.tech/tags/n9000.md>), [networking](<https://devfeed.tech/tags/networking.md>), [networks](<https://devfeed.tech/tags/networks.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [optics](<https://devfeed.tech/tags/optics.md>), [pluggable-optics](<https://devfeed.tech/tags/pluggable-optics.md>), [silicon-one](<https://devfeed.tech/tags/silicon-one.md>)

### AI overview

Cisco and NVIDIA are developing the USFP pluggable optics form factor for 400G SerDes lanes, supporting a path to higher-density 1.6 Tbps Ethernet fabrics. Cisco also outlined Silicon One G300 systems, a G302 top-of-rack switch, and the A100L for lower-cost management switching. The article discusses Cisco SONiC options and Live Protect runtime vulnerability shielding, with availability planned for 2027.

### Source excerpt

Cisco and NVIDIA team up announcing a new USFP form factor for 400G SerDes lanes. Cisco also has a few new Silicon One offerings The post New USFP Pluggable Optics Form Factor for Higher Speeds from Cisco and NVIDIA appeared first on ServeTheHome.

## Powering your Raspberry Pi Over Ethernet using PoE

DevFeed: [Powering your Raspberry Pi Over Ethernet using PoE](<https://devfeed.tech/articles/powering-your-raspberry-pi-over-ethernet-using-poe-84207.md>)

Original publisher: [Read original article](<https://blog.adafruit.com/2026/10/09/powering-your-raspberry-pi-over-ethernet-using-poe/>)

Author: Takara

Published: 2026-10-09T20:00:16Z

Content type: tutorial

Language: en

Sources: [Adafruit Industries](<https://devfeed.tech/sources/adafruit-industries-makers-hackers-artists-designers-and-engineers.md>)

Topics: [Raspberry Pi](<https://devfeed.tech/topics/raspberry-pi.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>)

Tags: [ethernet](<https://devfeed.tech/tags/ethernet.md>), [piday](<https://devfeed.tech/tags/piday.md>), [poe](<https://devfeed.tech/tags/poe.md>), [power](<https://devfeed.tech/tags/power.md>), [raspberry-pi](<https://devfeed.tech/tags/raspberry-pi.md>)

### AI overview

The article explains how Power over Ethernet can power a Raspberry Pi through its Ethernet connection. Raspberry Pi 3 boards support PoE with an add-on HAT, and the article notes that compatible switches can be an affordable option.

### Source excerpt

PiMyLifeUp shared how to power your Raspberry Pi using Power over Ethernet (PoE). The PoE standard is a neat way to power devices without needing to plug them in physically to a power point. Power is instead delivered over your ethernet connection, this is super useful if you have to run an Ethernet cable to [...]

## Arista steps up its Ethernet challenge to Nvidia's AI scale-up networking fabric

DevFeed: [Arista steps up its Ethernet challenge to Nvidia's AI scale-up networking fabric](<https://devfeed.tech/articles/arista-steps-up-its-ethernet-challenge-to-nvidia-s-ai-scale-up-networking-fabric-67038.md>)

Original publisher: [Read original article](<https://www.networkworld.com/article/4232036/arista-steps-up-its-ethernet-challenge-to-nvidias-ai-scale-up-networking-fabric.html>)

Author: Michael Cooney

Published: 2026-10-07T22:45:17Z

Content type: news

Language: en

Sources: [Network World](<https://devfeed.tech/sources/network-world.md>)

Topics: [AI Networking](<https://devfeed.tech/topics/ai-networking.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [NVLink](<https://devfeed.tech/topics/nvlink.md>), [InfiniBand](<https://devfeed.tech/topics/infiniband.md>), [Cisco](<https://devfeed.tech/topics/cisco.md>), [NVIDIA Research](<https://devfeed.tech/topics/nvidia-research.md>)

Tags: [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [artificial-intelligence-enterprise-routers-network-switches-networking-networking-devices](<https://devfeed.tech/tags/artificial-intelligence-enterprise-routers-network-switches-networking-networking-devices.md>), [enterprise-routers](<https://devfeed.tech/tags/enterprise-routers.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [network-switches](<https://devfeed.tech/tags/network-switches.md>), [networking](<https://devfeed.tech/tags/networking.md>), [networking-devices](<https://devfeed.tech/tags/networking-devices.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [scale](<https://devfeed.tech/tags/scale.md>)

### AI overview

Arista introduced Ethernet reference architectures and partnerships intended to support AI scale-up and scale-out networking as an alternative to proprietary interconnects such as Nvidia NVLink. Its plans include three rack designs and diagnostics software for validating and monitoring network components across operating environments.

### Source excerpt

Arista sees Ethernet as the scale-up networking fabric of choice for AI and continues to build up its alternative to proprietary interconnects such as Nvidia's NVLink technology. This week, Arista rolled out new Ethernet reference architectures as well as a raft of integration partnerships with industry players--including AMD, Arm, Broadcom, Meta, Microsoft, and Qualcomm--aimed at helping large enterprises and hyperscale customers make Ethernet-based scale-up and scale-out designs a reality. Arista will offer its 7060EX7 Series switches, along with fiber patch panels, integrated liquid-cooling manifolds, drip trays with active leak detection, rack management controllers, and power shelves with backup battery units, and combine that with software and other technologies from partner vendors. All are based on standard Ethernet technologies: "Scale-up based on ESUN Ethernet, and scale-out based on Ethernet with UEC features," Hardev Singh, vice president and general manager of Arista's cloud and AI group, told Network World. "Arista's positioning contrasts with Nvidia proprietary NVLink and InfiniBand stacks," Singh said. The Ethernet for Scale-Up Networking (ESUN) initiative was formed by the nonprofit Open Compute Project and promises to advance the networking technology to handle scale-up connectivity across accelerated AI infrastructure. ESUN backers include AMD, Arista, ARM, Broadcom, Cisco, HPE Networking, Marvell, Meta, Microsoft, Nvidia, OpenAI and Oracle. ESUN provides network operators flexibility across accelerators, switching silicon, optics, physical rack designs, and network software while retaining the high bandwidth, low latency, and reliable in-order delivery required for collective communication, according to Arista. The Ultra Ethernet Consortium (UEC), meanwhile, includes AMD, Arista, Broadcom, Cisco, Eviden, HPE, Intel, Meta, and Microsoft. Its charter is to bring together industry leaders to build a complete Ethernet-based communication stack architec

## Arista takes Ethernet into the rack, and the network becomes the backplane

DevFeed: [Arista takes Ethernet into the rack, and the network becomes the backplane](<https://devfeed.tech/articles/arista-takes-ethernet-into-the-rack-and-the-network-becomes-the-backplane-67060.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/10/07/arista-takes-ethernet-rack-network-becomes-backplane/>)

Author: Zeus Kerravala

Published: 2026-10-07T20:02:05Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [Local AI](<https://devfeed.tech/topics/local-ai.md>), [Cloud](<https://devfeed.tech/topics/cloud.md>), [Load Balancing](<https://devfeed.tech/topics/load-balancing.md>)

Tags: [advanced-micro-devices](<https://devfeed.tech/tags/advanced-micro-devices.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [amd](<https://devfeed.tech/tags/amd.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [arista](<https://devfeed.tech/tags/arista.md>), [arm](<https://devfeed.tech/tags/arm.md>), [artiicial-intelligence](<https://devfeed.tech/tags/artiicial-intelligence.md>), [aws](<https://devfeed.tech/tags/aws.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [broadcom](<https://devfeed.tech/tags/broadcom.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [infra](<https://devfeed.tech/tags/infra.md>), [network](<https://devfeed.tech/tags/network.md>), [networking](<https://devfeed.tech/tags/networking.md>), [news](<https://devfeed.tech/tags/news.md>), [nyse-anet](<https://devfeed.tech/tags/nyse-anet.md>), [switches](<https://devfeed.tech/tags/switches.md>), [the-latest](<https://devfeed.tech/tags/the-latest.md>)

### AI overview

Arista announced an open Ethernet rack-scale portfolio for AI infrastructure, including a scale-up architecture connecting up to 144 accelerators in one hop and up to 1,024 across racks. The article describes rack designs, cooling and integration requirements, and Arista's strategy to provide a common network stack across AI fabrics. It says adoption will depend on coordination across accelerator vendors, rack integrators and cooling suppliers.

### Source excerpt

Since its founding in 2004, it's arguable that no network vendor has made better switches than Arista Networks Inc. That was true then and remains true, as its prowess in this area helped it ride the cloud boom and the artificial intelligence wave. That model is changing as AI scales. Today, ahead of next week's [...] The post Arista takes Ethernet into the rack, and the network becomes the backplane appeared first on SiliconANGLE.

## NXP FRDM-IMXRT1186 development board features i.MX RT1186 crossover MCU, four Ethernet ports

DevFeed: [NXP FRDM-IMXRT1186 development board features i.MX RT1186 crossover MCU, four Ethernet ports](<https://devfeed.tech/articles/nxp-frdm-imxrt1186-development-board-features-i-mx-rt1186-crossover-mcu-four-ethernet-ports-61533.md>)

Original publisher: [Read original article](<https://www.cnx-software.com/2026/09/29/nxp-frdm-imxrt1186-development-board-features-i-mx-rt1186-crossover-mcu-four-ethernet-ports/>)

Author: Debashis Das

Published: 2026-09-29T09:00:20Z

Content type: news

Language: en

Sources: [CNX Software - Embedded Systems News](<https://devfeed.tech/sources/cnx-software-embedded-systems-news.md>)

Topics: [Microcontroller](<https://devfeed.tech/topics/microcontroller.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [Beta channel](<https://devfeed.tech/topics/beta-channel.md>), [Arm](<https://devfeed.tech/topics/arm.md>)

Tags: [arm](<https://devfeed.tech/tags/arm.md>), [automation](<https://devfeed.tech/tags/automation.md>), [development-board](<https://devfeed.tech/tags/development-board.md>), [edge-computing](<https://devfeed.tech/tags/edge-computing.md>), [embedded](<https://devfeed.tech/tags/embedded.md>), [embedded-systems](<https://devfeed.tech/tags/embedded-systems.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [gigabit-ethernet](<https://devfeed.tech/tags/gigabit-ethernet.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [industrial](<https://devfeed.tech/tags/industrial.md>), [mcu](<https://devfeed.tech/tags/mcu.md>), [nxp](<https://devfeed.tech/tags/nxp.md>), [nxp-i-mx](<https://devfeed.tech/tags/nxp-i-mx.md>), [plc](<https://devfeed.tech/tags/plc.md>), [rtos](<https://devfeed.tech/tags/rtos.md>), [video](<https://devfeed.tech/tags/video.md>), [zephyr](<https://devfeed.tech/tags/zephyr.md>), [zephyr-os](<https://devfeed.tech/tags/zephyr-os.md>)

### AI overview

NXP launched the FRDM-IMXRT1186, a development board built around the i.MX RT1186 crossover MCU for industrial automation, robotics, PLCs, and edge computing. Its dual-core design pairs an 800 MHz Cortex-M7 application core with a 300 MHz Cortex-M33 real-time control and communications core. The board provides four Ethernet ports, memory and flash, USB-C, CAN, debugger and JTAG connections, and expansion headers.

### Source excerpt

NXP has just launched the FRDM-IMXRT1186, a development board based on the i.MX RT1186 crossover MCU for industrial automation, robotics, PLCs, and edge computing applications. The MCU uses a heterogeneous architecture with an 800 MHz Arm Cortex-M7 core for application processing and a 300 MHz Arm Cortex-M33 core for real-time control and communications. The board includes 64 Mbit HyperRAM, 512 Mbit HyperFlash, and 128 Mbit QSPI Flash, along with two Gigabit Ethernet ports for TSN switching and two Fast Ethernet ports for EtherCAT or TSN switching. It also features a USB Type-C OTG port and a CAN transceiver, while debugging is supported through an on-board MCU-Link debugger and JTAG connector. For expansion, the board includes Arduino UNO R3 and FRDM headers, a FlexSPI follower header, an SRAMC header, and motor-control headers. NXP FRDM-IMXRT1186 development board specifications: SoC - NXP MIMXRT1186CVJ8C (i.MX RT1180 family, LFBGA196) CPU Arm Cortex-M7 application core @ [...] The post NXP FRDM-IMXRT1186 development board features i.MX RT1186 crossover MCU, four Ethernet ports appeared first on CNX Software - Embedded Systems News.

## WhiteFiber Continuum Goes Commercial, Scaling Its 83 km Two-Site GPU Supercluster Design to 136 Tbps

DevFeed: [WhiteFiber Continuum Goes Commercial, Scaling Its 83 km Two-Site GPU Supercluster Design to 136 Tbps](<https://devfeed.tech/articles/whitefiber-continuum-goes-commercial-scaling-its-83-km-two-site-gpu-supercluster-design-to-136-tbps-60031.md>)

Original publisher: [Read original article](<https://www.storagereview.com/news/whitefiber-continuum-goes-commercial-scaling-its-83-km-two-site-gpu-supercluster-design-to-136-tbps>)

Author: Harold Fritts

Published: 2026-09-25T14:56:39Z

Content type: news

Language: en

Sources: [StorageReview.com](<https://devfeed.tech/sources/storagereview-com.md>)

Topics: [GPU](<https://devfeed.tech/topics/gpu.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [data centers](<https://devfeed.tech/topics/data-centers.md>), [Network](<https://devfeed.tech/topics/network.md>), [Latency](<https://devfeed.tech/topics/latency.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [compute](<https://devfeed.tech/tags/compute.md>), [data-center](<https://devfeed.tech/tags/data-center.md>), [data-center-networking](<https://devfeed.tech/tags/data-center-networking.md>), [deploy](<https://devfeed.tech/tags/deploy.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [enterprises](<https://devfeed.tech/tags/enterprises.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [latency](<https://devfeed.tech/tags/latency.md>), [low-latency](<https://devfeed.tech/tags/low-latency.md>), [performance](<https://devfeed.tech/tags/performance.md>), [scaling](<https://devfeed.tech/tags/scaling.md>)

### AI overview

WhiteFiber has commercially launched Continuum, a distributed GPU supercluster that connects two QTS data centers 83 kilometers apart over dark fiber. The architecture uses DriveNets AI Fabric for networking and WEKA NeuralMesh for storage and memory, with a stated aggregate bandwidth of 136 Tbps and guaranteed round-trip latency of 0.9 milliseconds.

### Source excerpt

WhiteFiber has made WhiteFiber Continuum commercially available, turning the two-site GPU supercluster it first detailed as Project Redwood in July into a product enterprises can deploy. WhiteFiber calls Continuum its highest-performance cross-data-center networking solution and the first commercially available distributed GPU supercluster architecture. It links geographically separated data centers into a single logical GPU supercluster, The post WhiteFiber Continuum Goes Commercial, Scaling Its 83 km Two-Site GPU Supercluster Design to 136 Tbps appeared first on StorageReview.com.

## Review: M5 Ultra Mac Studio for AI, 3D, and Professional Video Workflows

DevFeed: [Review: M5 Ultra Mac Studio for AI, 3D, and Professional Video Workflows](<https://devfeed.tech/articles/review-m5-ultra-mac-studio-pure-unadulterated-power-57888.md>)

Original publisher: [Read original article](<https://www.computerworld.com/article/4225059/review-m5-ultra-mac-studio-pure-unadulterated-power.html>)

Author: Jonny Evans

Published: 2026-09-22T16:56:38Z

Content type: article

Language: en

Sources: [Computerworld](<https://devfeed.tech/sources/computerworld.md>)

Topics: [cpu](<https://devfeed.tech/topics/cpu.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Server](<https://devfeed.tech/topics/server.md>), [USB](<https://devfeed.tech/topics/usb.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [Wi-Fi](<https://devfeed.tech/topics/wi-fi.md>), [Bluetooth](<https://devfeed.tech/topics/bluetooth.md>), [Playback](<https://devfeed.tech/topics/playback.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [apple](<https://devfeed.tech/tags/apple.md>), [apple-artificial-intelligence-computers-and-peripherals-mac](<https://devfeed.tech/tags/apple-artificial-intelligence-computers-and-peripherals-mac.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [computers-and-peripherals](<https://devfeed.tech/tags/computers-and-peripherals.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [llms](<https://devfeed.tech/tags/llms.md>), [mac](<https://devfeed.tech/tags/mac.md>), [port](<https://devfeed.tech/tags/port.md>), [review](<https://devfeed.tech/tags/review.md>), [server](<https://devfeed.tech/tags/server.md>)

### AI overview

A review of Apple's M5 Ultra Mac Studio finds it substantially overpowered for most professional workflows. The article highlights its CPU and GPU performance, support for large locally run language models, display and video-production capabilities, connectivity, low-wattage operation, and limitations including its high price and non-upgradable memory and storage.

### Source excerpt

I'm old enough to remember when Macs were friendly little machines. While a little underpowered and equipped with some platform-unique foibles, they were really good at some things, highly secure, and had the best user interface of any PC. That was then; this is now. And while the platform is still unique, still highly secure, and still has the best user interface, Apple's all-new Mac Studio is a beast of a machine. It's incredibly powerful and would crackle with energy if it weren't so energy efficient. TL;DR review No matter what pro workflow you follow, the M5 Ultra Mac Studio is more than you need. Longer review Apple hasn't changed the design of the Mac Studio. It's 7.7-inch square, 3.7-inch high aluminum box with two USB-C and one SDXC port on the front, and an array of ports -- four Thunderbolt 5, two USB A, 10Gb Ethernet, HDMI, and headphone -- at rear. It brings Wi-Fi 7 and Bluetooth 6, supports up to eight displays with up to 6K resolution at 60Hz, or four Studio Display XDR units. Basically, this one box can power a bank of displays, and for those using it in a live video production environment, Apple has also introduced generator locking (a.k.a. genlock) over USB-C, which lets the on-display image synchronize precisely with a pro video camera for the most precise editing and playback you'll find. Pros and cons Pros Massive CPU/GPU power capable of handling the most demanding 3D and AI tasks. Neural accelerators and unified memory make it ideal for training and running large LLMs on-premises privately. Doubled read/write performance over the previous generation via a new storage controller and fast NAND. Inclusion of four ProRes accelerators (handling up to 33 streams of 8K) and genlock over USB-C for precise sync. High performance at low wattage; runs quiet and cool compared to high-end PC alternatives. Features Thunderbolt 5, Wi-Fi 7, and Bluetooth 6. Works exceptionally well as a headless AI or rendering server. Cons Extremely high retail price ($12,299

## The GenAI Boom Accelerates And Transforms Ethernet Switching

DevFeed: [The GenAI Boom Accelerates And Transforms Ethernet Switching](<https://devfeed.tech/articles/the-genai-boom-accelerates-and-transforms-ethernet-switching-55527.md>)

Original publisher: [Read original article](<https://www.nextplatform.com/connect/2026/09/20/the-genai-boom-accelerates-and-transforms-ethernet-switching/5297603>)

Author: Timothy Prickett Morgan

Published: 2026-09-20T10:53:28Z

Content type: news

Language: en

Sources: [The Next Platform: In-depth coverage of high end computing](<https://devfeed.tech/sources/the-next-platform-in-depth-coverage-of-high-end-computing.md>)

Topics: [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [Generative AI](<https://devfeed.tech/topics/generative-ai.md>), [Network](<https://devfeed.tech/topics/network.md>), [data centers](<https://devfeed.tech/topics/data-centers.md>), [Server](<https://devfeed.tech/topics/server.md>), [Routing (disambiguation)](<https://devfeed.tech/topics/routing.md>), [Cisco](<https://devfeed.tech/topics/cisco.md>), [huawei](<https://devfeed.tech/topics/huawei.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [cisco](<https://devfeed.tech/tags/cisco.md>), [connect](<https://devfeed.tech/tags/connect.md>), [datacenter](<https://devfeed.tech/tags/datacenter.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [genai](<https://devfeed.tech/tags/genai.md>), [huawei](<https://devfeed.tech/tags/huawei.md>), [idc](<https://devfeed.tech/tags/idc.md>), [networks](<https://devfeed.tech/tags/networks.md>), [router](<https://devfeed.tech/tags/router.md>), [routing](<https://devfeed.tech/tags/routing.md>), [server](<https://devfeed.tech/tags/server.md>), [switch](<https://devfeed.tech/tags/switch.md>)

### AI overview

The GenAI boom is driving strong growth in Ethernet switch and router markets, although component shortages may limit expansion and contribute to higher prices. IDC data cited in the article shows substantial second-quarter 2026 revenue growth for switches and routers, with demand extending across datacenter, edge, campus, enterprise, service-provider, and cloud networks.

### Source excerpt

The server market is exploding thanks to the GenAI boom, and the Ethernet switch and router markets ...

[Next page](<https://devfeed.tech/topics/ethernet.md?cursor=WyIyMDI2LTA5LTIwVDEwOjUzOjI4KzAwOjAwIiwgIjI5NzMzMzQ0LTA4YWEtNGE0Mi04ZDVmLWRlMDU4MGRjYjNhNyJd>)