# pcie Published articles for pcie. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## Lenovo ThinkAgile VX850 V4 Takes On In-Memory Virtualization With New Deploy Tiers and Four-Hypervisor Express Bundles DevFeed: [Lenovo ThinkAgile VX850 V4 Takes On In-Memory Virtualization With New Deploy Tiers and Four-Hypervisor Express Bundles]() Original publisher: [Read original article]() Author: Harold Fritts Published: 2026-09-17T16:43:22Z Content type: news Language: en Sources: [StorageReview.com]() Topics: [virtualization](), [Hardware](), [Deployment](), [hyper-v](), [Windows](), [Microsoft](), [Azure](), [nutanix](), [intel]() Tags: [azure](), [capacity](), [deployment](), [enterprise](), [hardware](), [hyper-v](), [intel](), [microsoft](), [nutanix](), [nvme](), [pcie](), [server](), [virtualization]() ### AI overview Lenovo added the four-socket ThinkAgile VX850 V4 to its virtualization lineup for demanding in-memory workloads. The launch also introduces three Infrastructure Deployment Services tiers and pre-validated Top Choice Express bundles covering Microsoft, Red Hat, SUSE, and Nutanix. Lenovo has not yet published the VX850 V4's own product specifications. ### Source excerpt Lenovo has added the ThinkAgile VX850 V4 to its virtualization lineup, a four-socket node it describes as purpose-built for highly virtualized and demanding in-memory workloads, and paired the launch with a three-tier deployment service framework and a set of pre-validated Top Choice Express bundles spanning Microsoft, Red Hat, SUSE, and Nutanix. The announcement follows Lenovo's The post Lenovo ThinkAgile VX850 V4 Takes On In-Memory Virtualization With New Deploy Tiers and Four-Hypervisor Express Bundles appeared first on StorageReview.com. ## Fujitsu Formally Announces Their MONAKA 144 Core CPU DevFeed: [Fujitsu Formally Announces Their MONAKA 144 Core CPU]() Original publisher: [Read original article]() Author: Michael Larabel Published: 2026-09-17T13:42:37Z Content type: news Language: en Sources: [Phoronix]() Topics: [cpu](), [Arm](), [armv9](), [Inference Performance](), [Sovereign AI](), [Hardware](), [pcie](), [benchmarking]() Tags: [arm](), [armv9](), [benchmarking](), [cpu](), [desktop-linux](), [fujitsu](), [hardware](), [inference-performance](), [linux-benchmarking](), [linux-hardware-benchmarks](), [linux-hardware-reviews](), [linux-how-to](), [linux-performance](), [linux-server-benchmarks](), [open-source-graphics](), [pcie](), [phoronix](), [phoronix-test-suite](), [sovereign-ai](), [ubuntu-benchmarks](), [ubuntu-hardware]() ### AI overview Fujitsu formally announced the MONAKA, a 144-core Arm-based server CPU designed for sovereign AI infrastructure. The 2nm CPU supports up to 3.8GHz, memory speeds up to 8800MT/s, matrix-operation acceleration, SVE2 vector operations, and claimed AI inference throughput twice that of other CPUs. MONAKA servers will support PCIe Gen6, with availability beginning in November and broader availability in Europe in 2027. ### Source excerpt For two years now we have been curious about the Fujitsu MONAKA CPU since some slides about this Arm-based server processor first appeared and they upstreamed compiler support in GCC 15. This week Fujitsu formally announced the MONAKA CPU and with availability beginning in November... ## Minisforum N5R NAS Revealed - An ARM DIY 5-Bay NAS DevFeed: [Minisforum N5R NAS Revealed - An ARM DIY 5-Bay NAS]() Original publisher: [Read original article]() Author: Rob Andrews Published: 2026-09-16T16:00:07Z Content type: news Language: en Sources: [NAS Compares]() Topics: [DIY NAS](), [Hardware](), [systems](), [Arm](), [cpu](), [NVMe]() Tags: [10gbe-nas](), [10gbe-networking](), [2-5gbe](), [2-5gbe-nas](), [24-7-nas](), [5-bay-nas](), [ai-agent-nas](), [ai-nas](), [arm](), [arm-nas](), [arm-nas-2026](), [bluetooth](), [bluetooth-5-3](), [bluetooth-5-3-nas](), [business-nas](), [cix-arm-cpu](), [cix-cp8180](), [diy-nas](), [ecc](), [ecc-memory-nas](), [ecc-nas](), [ethernet](), [home-nas](), [home-server](), [ifa-2026](), [ifa-2026-nas](), [local-ai-nas](), [low-power-nas](), [lpcamm2-nas](), [m-2-nvme-nas](), [miniscloud-os](), [minisforum-ifa-2026](), [minisforum-miniscloud-os](), [minisforum-n5-air](), [minisforum-n5-max](), [nas](), [nas-hardware](), [network-attached-storage-nas](), [pcie](), [pre-release](), [uncategorised](), [usb]() ### AI overview Minisforum revealed the N5R, a pre-release five-bay DIY NAS announced at IFA 2026. It uses the Arm-based CIX CP8180 platform and includes ECC-capable LPCAMM2 memory, five SATA drive bays, an M.2 NVMe slot, 10GbE and 2.5GbE networking, Wi-Fi 6E, Bluetooth 5.3, and PCIe expansion. Final memory capacity, software capabilities, and pricing remain unconfirmed. ### Source excerpt New ARM v9.2 Powered DIY NAS Revealed - The Minisforum N5R Minisforum has used IFA 2026 to reveal another addition to its growing NAS range, this time in the form of the N5R. At first glance, it follows the same general 5-bay desktop design language as the company's other recent N5 systems, including the removable [...] ## AMD Sends Out Linux Patches For Enabling SEV-TIO TDISP With PCIe 6.0 DevFeed: [AMD Sends Out Linux Patches For Enabling SEV-TIO TDISP With PCIe 6.0]() Original publisher: [Read original article]() Author: Michael Larabel Published: 2026-09-16T13:02:52Z Content type: news Language: en Sources: [Phoronix]() Topics: [Linux Kernel](), [pcie](), [trusted-execution-environment](), [Confidential Computing](), [virtualization](), [Security](), [Encryption](), [MERN]() Tags: [amd](), [arm](), [attestation](), [confidential-computing](), [cpu](), [desktop-linux](), [encryption](), [intel](), [kernel](), [linux](), [linux-benchmarking](), [linux-hardware-benchmarks](), [linux-hardware-reviews](), [linux-how-to](), [linux-kernel](), [linux-performance](), [linux-server-benchmarks](), [open-source-graphics](), [pcie](), [phoronix](), [phoronix-test-suite](), [processors](), [risc-v](), [security](), [ubuntu-benchmarks](), [ubuntu-hardware]() ### AI overview AMD submitted 17 Linux kernel patches to enable SEV-TIO TDISP for PCIe 6.0 and newer. The work is intended to secure direct I/O device assignment for confidential-computing environments, including AMD Secure Encrypted Virtualization guest VMs, and includes a common TEE Security Manager developed with Intel, Arm, and RISC-V. ### Source excerpt The newest Linux kernel patches out of AMD for enhancing the upstream support with the new AMD EPYC 9006 "Venice" processors is for enabling SEV-TIO TDISP that is supported with PCI Express 6.0 and beyond... ## AAEON MIX-PTLWV1 Panther Lake industrial mini-ITX motherboard offers quad DisplayPort, dual 10GbE, and more DevFeed: [AAEON MIX-PTLWV1 Panther Lake industrial mini-ITX motherboard offers quad DisplayPort, dual 10GbE, and more]() Original publisher: [Read original article]() Author: Jean-Luc Aufranc (CNXSoft) Published: 2026-09-16T12:00:59Z Content type: news Language: en Sources: [CNX Software - Embedded Systems News]() Topics: [Embedded Systems](), [intel](), [Intel Core](), [ddr5](), [M.2 NVMe](), [pcie](), [Graphics](), [Specifications]() Tags: [10gbe](), [2-5gbe](), [4g-lte](), [5g](), [aaeon](), [artificial-intelligence-ai](), [automotive](), [ddr5](), [graphics](), [hardware](), [industrial](), [intel-core](), [linux](), [m-2-nvme](), [mini-itx](), [motherboard](), [nvme](), [panther-lake](), [pcie](), [single-board-computer](), [smart-city](), [specifications](), [ubuntu](), [windows-11]() ### AI overview CNX Software reports on the AAEON MIX-PTLWV1, an industrial mini-ITX motherboard built around Intel Core Ultra Series 3 Panther Lake processors. It supports up to 128GB of DDR5 memory, NVMe storage, four DisplayPort outputs, optional wireless and cellular modules, multiple Ethernet ports including up to two 10GbE connections, serial interfaces, and PCIe expansion. ### Source excerpt AAEON MIX-PTLWV1 is an industrial mini-ITX motherboard based on Intel Core Ultra Series 3 "Panther Lake" processors with up to two 10GbE and two 2.5GbE RJ45 jacks, and four 4K-capable DisplayPort video outputs. The motherboard also supports up to 128GB DDR5 C/SO-DIMM memory, M.2 NVMe SSD storage, optional WiFi and Bluetooth, and/or 4G LTE/5G cellular via M.2 modules, a PCIe Gen4 x8 slot for expansion, six serial interfaces, 12V-24V wide DC input, and more. AAEON MIX-PTLWV1 specifications: SoC - Intel Panther Lake processor (PTL-H404/PTL-H484 families) with Intel Xe LPG graphics; TDP: 15 to 25W System Memory - Up to 128GB via 2x 262-pin DDR5 C/SO-DIMM sockets up to 6400MHz (H404 SKU) or 7200MHz (H484 SKU) Storage - M.2 2280 M-Key (PCIe Gen4 x4) socket for NVMe SSD Display Interfaces 4x DisplayPort 1.4 up to 4096 x 2160 @ 60Hz 40-pin LVDS/eDP co-layout connector for Default: 18/24 bit dual-channel LVDS up [...] The post AAEON MIX-PTLWV1 Panther Lake industrial mini-ITX motherboard offers quad DisplayPort, dual 10GbE, and more appeared first on CNX Software - Embedded Systems News. ## The Mac mini's compact design is influencing competing mini PCs DevFeed: [The Mac mini's compact design is influencing competing mini PCs]() Original publisher: [Read original article]() Author: Sarang Sheth Published: 2026-09-15T19:15:48Z Content type: opinion Language: en Sources: [Yanko Design]() Topics: [Mac Mini](), [pc](), [ASUS](), [ddr5](), [UEFI]() Tags: [amd](), [apple](), [asus](), [benchmark](), [ddr5](), [design](), [desktops](), [desktops-editorial-product-design-mac-mini-mini-pcs](), [editorial](), [mac-mini](), [mini-pc](), [mini-pcs](), [pc](), [pcie](), [product-design](), [review](), [uefi]() ### AI overview The article argues that Apple's Mac mini dimensions and compact aluminum design have influenced competing mini PCs from ASUS, Geekom, Links, Dell, Lenovo and MSI. It also discusses differences in hardware, thermals and firmware, including a UEFI bug that affected one Geekom benchmark. ### Source excerpt The Mac mini Form Factor is So Perfect, Literally Everyone is Copying It Apple's Mac mini is a 127 by 127 by 50mm aluminum square, and the PC industry has spent 2026 rebuilding that square with different logos... ## Astera Labs Releases Leo 2 CXL Memory Controllers and Leo X Controller for Rackscale Fabric-Attached Memory DevFeed: [Astera Labs Releases Leo 2 CXL Memory Controllers and Leo X Controller for Rackscale Fabric-Attached Memory]() Original publisher: [Read original article]() Author: Ryan Smith Published: 2026-09-15T17:00:48Z Content type: article Language: en Sources: [ServeTheHome]() Topics: [pcie](), [systems](), [datacenter]() Tags: [astera-labs](), [cxl](), [data-center](), [ddr4](), [ddr5](), [leo](), [memory](), [other-components](), [pcie](), [releases](), [server]() ### AI overview Astera Labs is launching Leo 2 CXL smart memory controllers with DDR4 support and CXL 3.2/PCIe Gen6 connectivity, alongside the Leo-X controller for fabric-attached memory in rack-scale systems. ### Source excerpt Astera Labs is launching a new generation of Leo smart memory controllers. The Leo 2 series adds support for CXL 3.2 and PCIe Gen6, while the ambitious Leo X brings the ability to attach memory expanders directly to the fabric networks of AI accelerators The post Astera Labs Releases Leo 2 CXL Memory Controllers and Leo X Controller for Rackscale Fabric-Attached Memory appeared first on ServeTheHome. ## Steam Frame vs. Quest 3 Specs: Better PC Streaming, Power & Hackability DevFeed: [Steam Frame vs. Quest 3 Specs: Better PC Streaming, Power & Hackability]() Original publisher: [Read original article]() Author: Scott Hayden Published: 2026-09-14T16:59:00Z Content type: comparison Language: en Sources: [Road to VR]() Topics: [Streaming](), [pc](), [Linux](), [Steam Deck](), [Windows](), [Operating system](), [Arm]() Tags: [arm](), [cameras](), [compatibility](), [devices](), [hacking](), [linux](), [meta-quest-3-news-reviews](), [news](), [os](), [pc](), [pc-vr-news-reviews](), [pcie](), [steam-deck](), [streaming](), [windows](), [x86]() ### AI overview This comparison examines Valve's Steam Frame standalone headset against Quest 3. Steam Frame is designed to run Steam's flatscreen and PC VR libraries, stream PC VR games wirelessly through a dedicated Wi-Fi 6E dongle, and support native x86 Windows and Linux content through a compatibility layer. Compared with Quest 3, it offers a more powerful Qualcomm Snapdragon processor, a more open and hackable SteamOS platform, an accessible gen4 PCIe expansion port, and broader compatibility, although its application compatibility is not yet as established. ### Source excerpt Want to know how Valve's Steam Frame standalone headset stacks up against the competition? Read on to find out. Steam Frame is a lot like Steam Deck; it can play pretty much all of Steam's game library out of the box, which includes flatscreen content based on x86 architecture--basically everything on Steam--and also now PC [...] The post Steam Frame vs. Quest 3 Specs: Better PC Streaming, Power & Hackability appeared first on Road to VR. ## QNAP TVS-hx77AX Brings NFS over RDMA and U.2 NVMe to Desktop ZFS NAS DevFeed: [QNAP TVS-hx77AX Brings NFS over RDMA and U.2 NVMe to Desktop ZFS NAS]() Original publisher: [Read original article]() Author: Lyle Smith Published: 2026-09-09T17:18:09Z Content type: news Language: en Sources: [StorageReview.com]() Topics: [systems](), [networking](), [virtualization](), [cpu](), [Network](), [GPU](), [Artificial Intelligence]() Tags: [10gbase-t](), [2-5gbe](), [amd](), [cpu](), [enterprise](), [gpu](), [medium-nas](), [memory](), [nas](), [networking](), [nvme](), [pcie](), [qnap](), [virtualization]() ### AI overview QNAP introduced the TVS-hx77AX series of desktop ZFS NAS systems, combining AMD Ryzen 7000 processors, DDR5 memory, U.2 NVMe storage, and NFS over RDMA. The 16-, 12-, and 8-bay models target video collaboration, virtualization, and small-scale AI workloads. ### Source excerpt QNAP has introduced the TVS-hx77AX series, a new line of desktop ZFS NAS systems combining AMD Ryzen 7000 series processors with DDR5 memory, U.2 NVMe storage, and NFS over RDMA support. The family includes the 16-bay TVS-h1677AX, 12-bay TVS-h1277AX, and 8-bay TVS-h877AX, and QNAP is aiming all three at video collaboration, virtualization, and small-scale AI The post QNAP TVS-hx77AX Brings NFS over RDMA and U.2 NVMe to Desktop ZFS NAS appeared first on StorageReview.com. ## Arm Neoverse CSS N4 Launched for Next-Gen CPUs and DPUs DevFeed: [Arm Neoverse CSS N4 Launched for Next-Gen CPUs and DPUs]() Original publisher: [Read original article]() Author: Cliff Robinson Published: 2026-09-09T17:10:12Z Content type: news Language: en Sources: [ServeTheHome]() Topics: [Arm](), [cpu](), [datacenter]() Tags: [architecture](), [arm](), [armv9](), [compute](), [connectivity](), [core](), [cpu](), [ddr5](), [memory](), [mesh](), [neoverse](), [neoverse-css](), [neoverse-n4](), [pcie](), [server](), [server-cpus]() ### AI overview Arm announced the Neoverse CSS N4 compute subsystem for future server CPUs and DPUs. The 3nm design supports 8 to 128 cores per die, Armv9.3, PCIe Gen7, LPDDR6, DDR5, MRDIMMs, and chiplet connectivity. ### Source excerpt The new Arm Neoverse CSS N4 IP is out so companies can quickly build next-generation power-efficient PCIe Gen7 CPUs The post Arm Neoverse CSS N4 Launched for Next-Gen CPUs and DPUs appeared first on ServeTheHome. ## KIOXIA CM9-R 15.36TB Review: BiCS8 Flash Hits Full Speed at Low Queue Depths DevFeed: [KIOXIA CM9-R 15.36TB Review: BiCS8 Flash Hits Full Speed at Low Queue Depths]() Original publisher: [Read original article]() Author: Brian Beeler Published: 2026-09-02T16:13:32Z Content type: article Language: en Sources: [StorageReview.com]() Topics: [datacenter](), [Security](), [Availability](), [configuration](), [.NET Framework](), [dell](), [intel](), [Nvidia]() Tags: [dell](), [enterprise](), [fips-140-3](), [high-availability](), [intel](), [lts](), [nvidia](), [pcie](), [performance](), [review](), [security](), [ssd](), [ubuntu]() ### AI overview This review evaluates the 15.36TB E3.S KIOXIA CM9-R enterprise SSD, built on eighth-generation BiCS FLASH. It highlights strong low-queue-depth performance, including 3,400K random read IOPS, 14,800 MB/s sequential read, 11,000 MB/s sequential write, and 540K random write IOPS, while emphasizing a typical active power rating of 25W. The drive supports PCIe 5.0 x4, dual-port operation, NVMe 2.0, NVMe-MI 1.2c, power-loss protection, end-to-end data protection, and optional security variants. ### Source excerpt The KIOXIA CM9-R is the first enterprise SSD family built on eighth-generation BiCS FLASH, and the 15.36TB E3.S model in our lab is the volume sweet spot of the range: 3,400K random read IOPS and 14,800 MB/s of sequential read from a drive with a 25W typical active rating. Gen5 enterprise drives routinely push into The post KIOXIA CM9-R 15.36TB Review: BiCS8 Flash Hits Full Speed at Low Queue Depths appeared first on StorageReview.com. ## Orange Pi Zero 4 - Compact Allwinner A733 SBC offers HDMI, USB-C DP, GbE, WiFI 6, PCIe FFC connector for $26.5 and up DevFeed: [Orange Pi Zero 4 - Compact Allwinner A733 SBC offers HDMI, USB-C DP, GbE, WiFI 6, PCIe FFC connector for $26.5 and up]() Original publisher: [Read original article]() Author: Jean-Luc Aufranc (CNXSoft) Published: 2026-08-31T07:13:46Z Content type: news Language: en Sources: [CNX Software - Embedded Systems News]() Topics: [Embedded Systems](), [SOC](), [Arm](), [cpu](), [wifi 6](), [GPU](), [Bluetooth](), [RISC-V](), [hdmi](), [USB](), [Ethernet](), [OpenGL]() Tags: [allwinner](), [allwinner-a-series](), [android](), [arm](), [armbian](), [bluetooth](), [cortex-a55](), [cpu](), [debian](), [embedded-systems](), [gigabit-ethernet](), [gpu](), [h-264](), [hardware](), [hdmi](), [linux](), [npu](), [opencl](), [orange-pi](), [pcie](), [single-board-computer](), [specifications](), [ubuntu](), [usb](), [vp9](), [vulkan](), [wifi-6]() ### AI overview The Orange Pi Zero 4 is a compact single-board computer based on the Allwinner A733, with up to 8GB of memory, eMMC or UFS storage support, mini HDMI, USB-C DisplayPort, Gigabit Ethernet, Wi-Fi 6, Bluetooth 5.4, camera interfaces, GPIO, and a PCIe FFC connector. The article lists its specifications, planned operating-system images, and expected Armbian support. ### Source excerpt Orange Pi Zero 4 is an affordable yet feature-rich Allwinner A733 single board computer (SBC) with up to 8GB RAM, eMMC/UFS storage footprint, a microSD card slot, and plenty of I/Os. The board packs mini HDMI and USB-C DisplayPort, Gigabit Ethernet, WiFi 6 and Bluetooth 5.4, two MIPI CSI camera connectors, a Raspberry Pi-style PCIe FFC connector, and two GPIO headers into a compact 55x50mm form factor. Orange Pi Zero 4 specifications: SoC - Allwinner A733 CPU Dual-core Arm Cortex-A76 @ up to 2.00 GHz Hexa-core Arm Cortex-A55 @ up to 1.79 GHz Single-core RISC-V E902 real-time core up to 200 MHz GPU - Imagination Technologies BXM-4-64 MC1 GPU with support for OpenGL ES 3.2, Vulkan 1.3, OpenCL 3.0 VPU 8Kp24 H.265/VP9/AVS2 decoding 4Kp30 H.265/H.264 encoding AI accelerator - 3 TOPS NPU System Memory - 1GB, 2GB, 4GB, 6GB, or 8GB LPDDR5/LPDDR4/LPDDR4x; note: 16GB is also listed in the specs, [...] The post Orange Pi Zero 4 - Compact Allwinner A733 SBC offers HDMI, USB-C DP, GbE, WiFI 6, PCIe FFC connector for $26.5 and up appeared first on CNX Software - Embedded Systems News. ## Hot Chips 2026: XCENA and Samsung's Near-Memory Compute CXL Device DevFeed: [Hot Chips 2026: XCENA and Samsung's Near-Memory Compute CXL Device]() Original publisher: [Read original article]() Author: Chester Lam Published: 2026-08-30T07:25:37Z Content type: article Language: en Sources: [Chips and Cheese]() Topics: [samsung](), [ddr5](), [RISC-V](), [data](), [Cache](), [Arm](), [intel]() Tags: [2026](), [arm](), [cache](), [capacity](), [clusters](), [compute](), [core](), [ddr5](), [dram](), [intel](), [memory](), [parallel](), [pcie](), [performance](), [risc-v](), [samsung]() ### AI overview The article examines XCENA and Samsung's MX1, a CXL memory expansion device that can host up to 2 TB of DDR5 memory, connect SSDs, and provide onboard compute through 3,072 RISC-V cores. It describes the device's memory bandwidth, cache hierarchy, power use, and focus on data-parallel workloads. ### Source excerpt CXL memory expansion, with a side of compute ## Recap: Zephyr Project Meetup - Sapporo, Japan 2026 DevFeed: [Recap: Zephyr Project Meetup - Sapporo, Japan 2026]() Original publisher: [Read original article]() Author: Zephyr Project Published: 2026-08-27T05:00:08Z Content type: article Language: en Sources: [Zephyr Project]() Topics: [Embedded Systems](), [Embedded Software Dev](), [Raspberry Pi](), [Open Source]() Tags: [2026](), [blog](), [boot](), [community](), [debugging](), [embedded](), [events](), [hardware](), [japan](), [meetup](), [member-blog](), [open-source](), [pcie](), [raspberry-pi-5](), [recap]() ### AI overview A recap of the June 26, 2026 Zephyr Project Meetup in Sapporo, Japan, covering community discussion, embedded engineering talks, and work to boot Zephyr on Raspberry Pi 5. ### Source excerpt This blog was written by members of the Zephyr community in Japan. The meetup speakers are referred to by their Discord names throughout the blog. On June 26, 2026, the Zephyr Project Meetup returned to Sapporo, Hokkaido, bringing together Zephyr users, embedded developers, hardware engineers, researchers, and open source contributors for an afternoon of technical talks and community discussion. ## Hot Chips 2026: CUDA Targets RISC-V DevFeed: [Hot Chips 2026: CUDA Targets RISC-V]() Original publisher: [Read original article]() Author: Chester Lam Published: 2026-08-24T16:49:25Z Content type: article Language: en Sources: [Chips and Cheese]() Topics: [CUDA](), [RISC-V](), [Nvidia](), [GPU](), [Software]() Tags: [cache](), [cuda](), [framework](), [gpu](), [nvidia](), [pcie](), [risc-v](), [software](), [uefi]() ### AI overview The article reports that Nvidia is exploring support for CUDA on RISC-V CPUs. It describes requirements including RVA23, server SoC and platform specifications, RISC-V ACPI support, vector extensions, and PCIe coherency to support reliable and efficient GPU compute. ### Source excerpt CUDA is the most important software framework in the GPU compute world, and Nvidia is looking at supporting CUDA on RISC-V. Terms and conditions may apply. ## Self-healing GPU nodes in Kubernetes: What we learned building the EKS node monitoring agent DevFeed: [Self-healing GPU nodes in Kubernetes: What we learned building the EKS node monitoring agent]() Original publisher: [Read original article]() Author: Sajjan Gundapuneedi Published: 2026-07-19T13:00:00Z Content type: article Language: en Sources: [Kubernetes Overview, News and Trends | The New Stack]() Topics: [Kubernetes](), [Amazon EKS](), [GPU](), [Monitoring]() Tags: [ai-infrastructure](), [amazon-eks](), [aws-marketplace](), [gpu](), [kubernetes](), [monitoring](), [open-source](), [pcie](), [post-contributed](), [self-healing](), [sponsor-aws-marketplace](), [sponsored-post-contributed]() ### AI overview The article describes how the EKS Node Monitoring Agent detects Kubernetes node failures, writes NodeConditions, and works with Karpenter to replace affected nodes. It explains design lessons from operating automatic repair at GPU scale and places the agent within EKS Auto Mode. ### Source excerpt When you run Kubernetes at the scale we do on Amazon EKS, nodes break constantly. GPUs fall off the PCIe The post Self-healing GPU nodes in Kubernetes: What we learned building the EKS node monitoring agent appeared first on The New Stack. ## Upcoming Livestream: GPUs for AI (Shaped by You) DevFeed: [Upcoming Livestream: GPUs for AI (Shaped by You)]() Original publisher: [Read original article]() Author: Alex Razvant Published: 2026-01-27T09:30:49Z Content type: article Language: en Sources: [Neural Bits]() Topics: [GPU](), [Artificial Intelligence](), [Hardware](), [CUDA](), [pcie](), [Nvidia](), [rocm](), [MLX](), [TensorRT](), [Google](), [groq]() Tags: [ai](), [cuda](), [google](), [gpu](), [groq](), [hardware](), [mlx](), [nvidia](), [pcie](), [rocm](), [tensorrt]() ### AI overview An upcoming livestream will discuss how GPUs and other accelerators support AI workloads. The author invites audience feedback to shape coverage of GPU hardware, PCIe, CUDA, ASICs, TPUs, LPUs, model optimization, and related technologies. ### Source excerpt You can choose the topics for a Live Session on GPUs in AI ## An AI Engineer's Guide To Choosing GPUs DevFeed: [An AI Engineer's Guide To Choosing GPUs]() Original publisher: [Read original article]() Author: Alex Razvant Published: 2025-12-07T14:02:40Z Content type: tutorial Language: en Sources: [Neural Bits]() Topics: [GPU](), [Nvidia](), [Hardware](), [Machine learning](), [Artificial Intelligence](), [Inference](), [Large Language Model](), [Blackwell](), [Hopper](), [lora](), [NCCL](), [Kernel]() Tags: [ai](), [ai-engineer](), [architecture](), [blackwell](), [clusters](), [compute](), [data-center](), [deep-dive](), [gpu](), [guide](), [hardware](), [hopper](), [inference](), [llm](), [ml](), [nvidia](), [nvlink](), [pcie](), [software](), [technical]() ### AI overview A technical guide to choosing NVIDIA GPUs for AI workloads. It explains how GPU microarchitecture, memory subsystems, form factors, and interconnects affect capabilities, scaling, training, and inference, and compares consumer and data-center GPUs. ### Source excerpt A deep dive on technical Hardware and Software details of NVIDIA GPUs for AI Workloads. ## A real-world comparison between Lenovo's ThinkBook 16p Gen 2 and 16p Gen 6 (with photos!) DevFeed: [A real-world comparison between Lenovo's ThinkBook 16p Gen 2 and 16p Gen 6 (with photos!)]() Original publisher: [Read original article]() Author: Jake Lee Published: 2025-12-03T00:00:00Z Content type: comparison Language: en Sources: [Jake Lee's Programming Blog]() Topics: [lenovo](), [Benchmark](), [3dmark](), [GPU](), [cpu](), [Hardware](), [pcie](), [Programming]() Tags: [3dmark](), [benchmarks](), [comparison](), [cpu](), [gpu](), [hardware](), [laptop](), [lenovo](), [pcie](), [performance](), [technology](), [tests]() ### AI overview A real-world comparison of the Lenovo ThinkBook 16p Gen 2 and Gen 6 examines specifications, photos, benchmarks, storage performance, and gaming results. The author reports substantial improvements in SSD and 3DMark performance, while noting that benchmark results may not fully represent everyday use. ### Source excerpt I recently replaced my aging Lenovo ThinkBook 16p Gen 2 with a Gen 6, so here's benchmarks, side-by-side photos, and general notes on their differences! ## Expanding a Home Server with OCuLink NVMe Storage and Comparing Samsung 990 Pro with Intel Optane DevFeed: [Expanding a Home Server with OCuLink NVMe Storage and Comparing Samsung 990 Pro with Intel Optane]() Original publisher: [Read original article]() Author: Techno Tim Published: 2025-04-26T13:00:00Z Content type: tutorial Language: en Sources: [Techno Tim]() Topics: [Homelab](), [Hardware](), [Latency](), [cloud-infrastructure](), [intel](), [samsung](), [servers]() Tags: [hardware](), [home-server](), [homelab](), [intel](), [latency](), [pcie](), [samsung](), [server](), [tests]() ### AI overview This tutorial documents expanding a home server with additional NVMe storage through OCuLink. It explains how to inspect NVMe drives, PCIe lane assignments, negotiated speeds, SMART data, and benchmark results while comparing a Samsung 990 Pro with an Intel Optane P1600X. The reported results show lower latency and higher IOPS for the Optane drive, while the Samsung drive achieves higher raw transfer speeds. ### Source excerpt I thought I had no PCIe lanes left -- but OCuLink opened a new path. 📺 Watch Video Info Today, I expand my home server with more NVMe storage, put a Samsung 990 Pro and Intel Optane head-to-head, and show just how much you can still get out of a packed system. Tests I decided to test each drive to be sure that I wasn't being throttled and that I was getting the highest speeds my motherboa... ## HomeLab Hardware Tour (Early 2025) DevFeed: [HomeLab Hardware Tour (Early 2025)]() Original publisher: [Read original article]() Author: Techno Tim Published: 2025-01-06T13:00:00Z Content type: article Language: en Sources: [Techno Tim]() Topics: [Hardware](), [Homelab](), [Server](), [Network]() Tags: [2025](), [ddr4-so-dimm](), [hardware](), [homelab](), [network](), [nvme](), [pcie](), [pikvm](), [samsung](), [servers]() ### AI overview A 2025 tour of the hardware used in Techno Tim's homelab, covering rack equipment, networking devices, servers, storage, UPS units, smart-home hardware, and related accessories. ### Source excerpt Here's all of the hardware I am running in my Homelab in 2025! Software / services tour soon! 📺 Watch Video Info Merch Shop 🛍: https://l.technotim.com/shop 📦 Gear in this video 📦 Rack & Accessories 32u Rack https://amzn.to/3TlhdXb D Ring Hooks https://amzn.to/3FSxk5Y 1u Brush Panels https://amzn.to/3hKNiaf 1u Rack Mount Full Depth Shelf https://amzn.to/3jrm5Kp Right Ang... ## Accelerating Protein Language Model ProtST on Intel Gaudi 2 DevFeed: [Accelerating Protein Language Model ProtST on Intel Gaudi 2]() Original publisher: [Read original article]() Author: Julien Simon; Jiqing.Feng; Santiago Miret; Xinyu Yuan; Yi Wang; Matrix Yao; Minghao Xu; Ke Ding Published: 2024-07-03T00:00:00Z Content type: tutorial Language: en Sources: [Hugging Face - Blog]() Topics: [Large Language Model](), [Inference](), [Inference Performance](), [Fine-tuning](), [intel](), [Open Source](), [AI, ML & Data Engineering]() Tags: [accelerators](), [artificial-intelligence](), [batch](), [datasets](), [fine-tuning](), [hugging-face](), [inference](), [inference-performance](), [intel](), [language-models](), [llm](), [model](), [nvidia](), [optimum](), [partnerships](), [pcie](), [precision](), [transformers](), [zero-shot]() ### AI overview This tutorial explains how to run inference and fine-tune ProtST, a multimodal protein language model, using Intel Gaudi 2 accelerators and the Optimum for Intel Gaudi open-source library. It compares ProtST inference on NVIDIA A100 and Gaudi 2, reporting identical accuracy and 1.76x faster inference on Gaudi 2. ### Source excerpt We're on a journey to advance and democratize artificial intelligence through open source and open science. ## SSD Performance Has Advanced Faster Than Cloud Vendor NVMe Instances DevFeed: [SSD Performance Has Advanced Faster Than Cloud Vendor NVMe Instances]() Original publisher: [Read original article]() Author: Viktor Leis (noreply@blogger.com) Published: 2024-02-19T08:00:00Z Content type: opinion Language: en Sources: [Database Architects]() Topics: [Cloud](), [Amazon EC2](), [NVMe](), [pcie](), [datacenter]() Tags: [aws](), [capacity](), [cloud](), [cost](), [data-center](), [ec2](), [io](), [nvme](), [pcie](), [performance](), [server](), [storage]() ### AI overview The article compares rapid advances in commodity PCIe and NVMe SSD performance with slower progress in cloud-based NVMe instances. It argues that AWS EC2 instance performance per SSD has remained around 2 GB/s since the i3 launch, leaving a substantial gap between leading data center SSDs and those offered by major cloud vendors. ### Source excerpt In recent years, flash-based SSDs have largely replaced disks for most storage use cases. Internally, each SSD consists of many independent flash chips, each of which can be accessed in parallel. Assuming the SSD controller keeps up, the throughput of an SSD therefore primarily depends on the interface speed to the host. In the past six years, we have seen a rapid transition from SATA to PCIe 3.0 to PCIe 4.0 to PCIe 5.0. As a result, there was an explosion in SSD throughput: At the same time, we saw not just better performance, but also more capacity per dollar: The two plots illustrate the power of a commodity market. The combination of open standards (NVMe and PCIe), huge demand, and competing vendors led to great benefits for customers. Today, top PCIe 5.0 data center SSDs such as the Kioxia CM7-R or Samsung PM1743 achieve up to 13 GB/s read throughput and 2.7M+ random read IOPS. Modern servers have around 100 PCIe lanes, making it possible to have a dozen of SSDs (each usually using 4 lanes) in a single server at full bandwidth. For example, in our lab we have a single-socket Zen 4 server with 8 Kioxia CM7-R SSDs, which achieves 100GB/s (!) I/O bandwidth: AWS EC2 was an early NVMe pioneer, launching the i3 instance with 8 physically-attached NVMe SSDs in early 2017. At that time, NVMe SSDs were still expensive, and having 8 in a single server was quite remarkable. The per-SSD read (2 GB/s) and write (1 GB/s) performance was considered state of the art as well. Another step forward occurred in 2019 with the launch of i3en instances, which doubled storage capacity per dollar. Since then, several NVMe instance types, including i4i and im4gn, have been launched. Surprisingly, however, the performance has not increased; seven years after the i3 launch, we are still stuck with 2 GB/s per SSD. Indeed, the venerable i3 and i3en instances basically remain the best EC2 has to offer in terms of IO-bandwidth/$ and SSD-capacity/$, respectively. Personally, I find this very s ## Booting the Raspberry Pi 5 from NVMe DevFeed: [Booting the Raspberry Pi 5 from NVMe]() Original publisher: [Read original article]() Author: Alex Ellis Published: 2023-12-28T17:55:43Z Content type: tutorial Language: en Sources: [Alex Ellis' Blog]() Topics: [NVMe](), [Raspberry Pi](), [boot](), [Linux](), [pcie](), [Ethernet](), [Edge](), [Google](), [Machine learning]() Tags: [boot](), [edge](), [ethernet](), [google](), [hdmi](), [linux](), [ml](), [mount](), [network](), [nvme](), [pcie](), [raspberry-pi](), [raspberry-pi-5](), [raspberry-pi-os]() ### AI overview A practical workflow for configuring a Raspberry Pi 5 to boot Raspberry Pi OS from NVMe for headless use. The article discusses first-generation PCIe breakout boards, recommends flashing a fresh image instead of cloning a booted SD card, and outlines setup steps using Ethernet, HDMI, or network scanning. ### Source excerpt Here's my workflow for setting up the Raspberry Pi 5 to boot from NVMe for headless use. [Next page]()