During the Vision 2024 event, Intel announced its latest Gaudi 3 AI accelerator, promising significant improvements over its predecessor. Intel claims the Gaudi 3 offers up to 70% improvement in training performance, 50% better inference, and 40% better efficiency than Nvidia's H100 processors. The new AI accelerator is presented as a PCIe Gen 5 dual-slot add-in card with a 600 W TDP or an OAM module with 900 W. The PCIe card has the same peak 1,835 TeraFLOPS of FP8 performance as the OAM module despite a 300 W lower TDP. The PCIe version works as a group of four per system, while the OAM HL-325L modules can be run in an eight-accelerator configuration per server. This likely will result in a lower sustained performance, given the lower TDP, but it confirms that the same silicon is used, just finetuned with a lower frequency. Built on TSMC's N5 5 nm node, the AI accelerator features 64 Tensor Cores, delivering double the FP8 and quadruple FP16 performance over the previous generation Gaudi 2.

The Gaudi 3 AI chip comes with 128 GB of HBM2E with 3.7 TB/s of bandwidth and 24 200 Gbps Ethernet NICs, with dual 400 Gbps NICs used for scale-out. All of that is laid out on 10 tiles that make up the Gaudi 3 accelerator, which you can see pictured below. There is 96 MB of SRAM split between two compute tiles, which acts as a low-level cache that bridges data communication between Tensor Cores and HBM memory. Intel also announced support for the new performance-boosting standardized MXFP4 data format and is developing an AI NIC ASIC for Ultra Ethernet Consortium-compliant networking. The Gaudi 3 supports clusters of up to 8192 cards, coming from 1024 nodes comprised of systems with eight accelerators. It is on track for volume production in Q3, offering a cost-effective alternative to NVIDIA accelerators with the additional promise of a more open ecosystem. More information and a deeper dive can be found in the Gaudi 3 Whitepaper.

ASUS today announced the RT-BE88U, the world's first WiFi 7 router that offers up to 34 Gbps of combined wired network capacity. With a 10 Gbps LAN / WAN port, a 10 Gbps LAN / WAN SFP+ port and four 2.5 Gbps ports, the RT-BE88U stands ready to deliver massive amounts of Ethernet performance for online gaming, media streaming and more - now and well into the future. It also offers up to 7200 Mbps of cutting-edge WiFi 7 networking, bringing the enhanced data rates of 4K-QAM and dual-band Multi-Link Operation to compatible devices. Plus, Guest Network Pro, subscription-free security features and site-to-site VPN support help to make an RT-BE88U-based network flexible, safe and easy to manage.

The WiFi 7 advantage
With WiFi 7 cornerstones such as the boosted data rates of 4K-QAM and dual-band Multi-Link Operation (MLO), the ASUS RT-BE88U ushers in a whole new wave of networking possibilities. 4K-QAM increases speeds by packing signals more tightly, which increases bandwidth efficiency. Thus, data rates can be up to 20% higher than what a WiFi 6 network can achieve. Meanwhile, Dual-band Multi-Link Operation (MLO) lets a WiFi 7 device connect to both a 2.4 GHz and a 5 GHz network to increase throughput and minimize latency. This future-proofs the network, giving users the capacity to handle even more connected devices without sacrificing performance

Dell Technologies is strengthening its collaboration with NVIDIA to help enterprises adopt AI technologies. By expanding the Dell Generative AI Solutions portfolio, including with the new Dell AI Factory with NVIDIA, organizations can accelerate integration of their data, AI tools and on-premises infrastructure to maximize their generative AI (GenAI) investments. "Our enterprise customers are looking for an easy way to implement AI solutions—that is exactly what Dell Technologies and NVIDIA are delivering," said Michael Dell, founder and CEO, Dell Technologies. "Through our combined efforts, organizations can seamlessly integrate data with their own use cases and streamline the development of customized GenAI models."

"AI factories are central to creating intelligence on an industrial scale," said Jensen Huang, founder and CEO, NVIDIA. "Together, NVIDIA and Dell are helping enterprises create AI factories to turn their proprietary data into powerful insights."

The HPC research projects DEEP-SEA, IO-SEA and RED-SEA are wrapping up this month after a three-year project term. The three projects worked together to develop key technologies for European Exascale supercomputers, based on the Modular Supercomputing Architecture (MSA), a blueprint architecture for highly efficient and scalable heterogeneous Exascale HPC systems. To achieve this, the three projects collaborated on system software and programming environments, data management and storage, as well as interconnects adapted to this architecture. The results of their joint work will be presented at a co-design workshop and poster session at the EuroHPC Summit (Antwerp, 18-21 March, www.eurohpcsummit.eu).

Marvell Technology, Inc., a leader in data infrastructure semiconductor solutions, is extending its collaboration with TSMC to develop the industry's first technology platform to produce 2 nm semiconductors optimized for accelerated infrastructure.

Behind the Marvell 2 nm platform is the company's industry-leading IP portfolio that covers the full spectrum of infrastructure requirements, including high-speed long-reach SerDes at speeds beyond 200 Gbps, processor subsystems, encryption engines, system-on-chip fabrics, chip-to-chip interconnects, and a variety of high-bandwidth physical layer interfaces for compute, memory, networking and storage architectures. These technologies will serve as the foundation for producing cloud-optimized custom compute accelerators, Ethernet switches, optical and copper interconnect digital signal processors, and other devices for powering AI clusters, cloud data centers and other accelerated infrastructure.

Senao Networks, a leading network solution provider, proudly announces its launch of SX904 SmartNIC based on the Intel NetSec Accelerator Reference Design. This cutting-edge NIC, harnessing the power of PCIe Gen 4 technology and fueled by the Intel Xeon D processor, sets an unprecedented standard in high-performance network computing. Senao will showcase a system demonstration at the Intel booth during the upcoming MWC in Barcelona. As transformative shift in networking edge, enterprises are increasingly leaning on scalable edge infrastructure. In order to cater the demands of workloads, low latency, local data processing, and robust security, SX904 marks a significant leap forward.

The combination of Intel XeonD processor, PCIe Gen 4 technology, dual 25 Gbps SFP28 support, and DDR4 ECC memory support enables the SX904 to achieve unparalleled data transfer rates and maximum bandwidth utilization, ideal for modern server architectures. It provides higher performance from the latest Intel Xeon D processor and Intel Ethernet Controller E810 and supports the latest Intel Platform Firmware Resilience, BMC, and TPM 2.0. SX904 enables the seamless offload of applications optimized for Intel architecture with zero changes, optimizing performance transmission effortlessly into an Intel-based server in PCIe add-in-card form factor.

Semtech Corporation, a high-performance semiconductor, IoT systems, and cloud connectivity service provider, today announces the AirLink XR60 5G Router Solution, a breakthrough in networking technology designed to offer 5G and Wi-Fi 6 performance in an ultra-compact and ultra-rugged design for highly demanding environments.

"Celebrating this release is particularly special for us at Semtech, formerly Sierra Wireless. The AirLink XR60 isn't just any router - it's the next evolution in 5G technology. Rising 5G adoption is paving the way for a more diverse array of applications, underlining the need for versatile and easy-to-install solutions. Our customers, particularly those in public safety and utilities, have expressed a clear requirement for routers that aren't just technologically advanced, but are also small, rugged and reliable under extreme conditions. The XR60 embodies our continued commitment to these sectors, ensuring that we're not only meeting but exceeding the expectations and demands of those who rely on us the most," explains Tom Mueller, EVP and GM of IoT System Products Group at Semtech.

This week, Cisco and NVIDIA have announced plans to deliver AI infrastructure solutions for the data center that are easy to deploy and manage, enabling the massive computing power that enterprises need to succeed in the AI era. "AI is fundamentally changing how we work and live, and history has shown that a shift of this magnitude is going to require enterprises to rethink and re-architect their infrastructures," said Chuck Robbins, Chair and CEO, Cisco. "Strengthening our great partnership with NVIDIA is going to arm enterprises with the technology and the expertise they need to build, deploy, manage, and secure AI solutions at scale." Jensen Huang, founder and CEO of NVIDIA said: "Companies everywhere are racing to transform their businesses with generative AI. Working closely with Cisco, we're making it easier than ever for enterprises to obtain the infrastructure they need to benefit from AI, the most powerful technology force of our lifetime."

A Powerful Partnership
Cisco, with its industry-leading expertise in Ethernet networking and extensive partner ecosystem, together with NVIDIA, the inventor of the GPU that fueled the AI boom, share a vision and commitment to help customers navigate the transitions for AI with highly secure Ethernet-based infrastructure. Cisco and NVIDIA have offered a broad range of integrated product solutions over the past several years across Webex collaboration devices and data center compute environments to enable hybrid workforces with flexible workspaces, AI-powered meetings and virtual desktop infrastructure.

Lenovo has released a pair of new ThinkVision displays, namely the P27pz-30 and the P32pz-30. These office monitors not only have 1,152 local dimming zones and a peak brightness of 1,200 nits, but they also incorporate a USB4 input. Somewhat unusually for Mini LED monitors, Lenovo has gone for IPS panels and both models sport a 3840 x 2160 resolution, a typical brightness of 650 nits, DisplayHDR 1000 certification, a contrast ratio of 1000:1 and a 4 or 6 ms response time depending on mode. There's also support for 99 percent Adobe RGB and 98 percent DCI-P3 colour gamut, although the panel is still an 8-bit+FRC panel and not a true 10-bit panel. The downside you ask? A 60 Hz refresh rate, which almost feels like a crime, despite the fact that these are productivity focused displays.

As for connectivity there's little lacking as besides the USB4 input which also supports DP Alt mode and 140 W USB PD, both displays also sport two HDMI 2.1 and one DP 1.4 input and one DP 1.4 output. Furthermore there is one USB Type-C 3.2 Gen 2 input, one USB Type-C 3.2 Gen 2 output with 15 W USB PD support, four USB Type-A 3.2 Gen 2 ports, one Gigabit Ethernet port and a 3.5 mm audio output. The USB4 solution appears to be based on VLI's VL830 or VL832 endpoint devices due to lack of any kind of USB4 output. Both models also feature built-in KVM functionality, a smart sensor and an internal power supply, as well as a height adjustable stand that also tilts and pivots. The kicker is the asking price, as Lenovo wants no less than €1,699 for the 27-inch P27pz-30 and €1,999 for the 32-inch P32pz-30.

As the Philips 49B2U6900CH makes its debut, it brings a plethora of new features along with it that increase collaboration and aid teamwork in the modern digital workplace. In addition to these new features, the Philips 49B2U6900CH also boasts a large 48.8-inch screen and 5120 x 1440 DQHD resolution for a wide viewing and high-quality setup that would otherwise be comprised of two separate monitors.

Built for the Professional Work Environment
Perhaps one of the most powerful and useful features of the Philips 49B2U6900CH are those that pertain to the professional work environment, and it is safe to say that it is built for it.

Intel Corporation today reported fourth-quarter and full-year 2023 financial results. "We delivered strong Q4 results, surpassing expectations for the fourth consecutive quarter with revenue at the higher end of our guidance," said Pat Gelsinger, Intel CEO. "The quarter capped a year of tremendous progress on Intel's transformation, where we consistently drove execution and accelerated innovation, resulting in strong customer momentum for our products. In 2024, we remain relentlessly focused on achieving process and product leadership, continuing to build our external foundry business and at-scale global manufacturing, and executing our mission to bring AI everywhere as we drive long-term value for stakeholders."

David Zinsner, Intel CFO, said, "We continued to drive operational efficiencies in the fourth quarter, and comfortably achieved our commitment to deliver $3 billion in cost savings in 2023. We expect to unlock further efficiencies in 2024 and beyond as we implement our new internal foundry model, which is designed to drive greater transparency and accountability and higher returns on our owners' capital." For the full year, the company generated $11.5 billion in cash from operations and paid dividends of $3.1 billion.

TP-Link, a leading global provider of consumer and business networking products, and the first-to-market Wi-Fi 7 brand, is proud to announce its latest line of Wi-Fi 7 innovations at CES 2024. This exciting lineup features a more robust collection of Deco Mesh Wi-Fi systems, next-gen Archer routers, and the industry's most powerful gaming router, along with the debut of the world's first Wi-Fi 7 client adapter. TP-Link is committed to delivering an unparalleled connectivity experience for homes of any size and internet access type.

Wi-Fi 7 Whole Home Mesh All-Scenario Solution
Deco offers a seamlessly integrated and comprehensive mesh Wi-Fi solution, allowing users to effortlessly create a whole home mesh Wi-Fi system. Whether consumers have a small apartment or a spacious residence with expansive floor plans and outdoor requirements, Deco ensures a robust and reliable network with faster speeds, lower latency, and more capacity than ever before. At CES 2024, TP-Link will unveil an impressive end-to-end range of Deco solutions that fit any scenario, including the BE95, BE65 Pro, BE63, BE65-5G, BE65-PoE, BE25, BE25-Outdoor, and more.

Razer also unveiled its new 11-port USB C Dock at CES 2024 show, and while it might not be interesting as the new Blade laptops or high-end gaming chairs, Razer is keeping up with the theme, making a simple USB C Dock feel like a premium device with its lightweight cnc aluminium chassis.

Razer USB C Dock comes with a total of 11 ports, including two standard USB 2.0 Type-A ports, two USB 3.2 Type-A ports, one standard USB 3.2 Type-C port and one USB Type-C Power Delivery port with support for 20 V USB PD 3.0 charging of up to 100 W, HDMI port with support for 4K @ 60 Hz output, Gigabit Ethernet RJ-45 port, microSD UHS-I and SD UHS-I card slots, and 3.5 mm audio combo jack powered by Razer 7.1 Surround Sound. The Razer USB C Dock is now available in Black and Mercury color options and priced at $119.99.

Today marks the launch of not just two new incredibly well-designed NAS devices, but also two new features brand new to the ASUSTOR line of NAS devices with ARM processors. The introduction of the Drivestor 2 Pro Gen2 and Drivestor 4 Pro Gen2 come with an upgraded SoC, providing 21% more performance, which increases read and write speeds and performance for the features and apps available for the Drivestor Pro Gen2. The Drivestor Pro Gen2 series comes with 2.5-Gigabit Ethernet, hot-swappable hard drive bays, hardware transcoding, tool-free design and more! The upgraded iGPU found in the Drivestor Pro Gen2 now gives even better performance when transcoding multimedia.

The all-new Drivestor Pro Gen2 series comes with a new killer feature. Btrfs is now supported on the Drivestor Pro Gen2, a first for ASUSTOR NAS devices running ARM CPUs. Btrfs helps protect data by enabling snapshots of data at certain points in time and is able to rewind the clock should any unintentional modifications be made and restore data if needed.

Raising the bar and becoming a CES 2024 Innovation Award Honoree, Dell Technologies is introducing the world's first 40-inch 5K monitor certified for five-star eye comfort, the Dell UltraSharp 40 Curved Thunderbolt Hub Monitor (U4025QW). Marking nine years of being an industry leader in PC monitors, Dell continues to push the boundaries of display innovation, bridging the gap for those who demand ultimate performance with the best technology and design.

Whether you're a content creator, data scientist or engineer, or you just need a monitor to keep you in your flow—a device with the best value in precision and efficiency is critical to keeping you productive in any work environment. Known for outstanding visuals, the U4025QW is a must-have for professionals who require clarity and remarkable color. This ultrawide curved VESA DisplayHDR 600 certified monitor comes with IPS Black Panel technology for greater color contrast and intricate detail in 5K resolution (5120x2160).

TerraMaster, a professional brand focused on providing innovative storage products for homes and businesses, recently introduced the 424 series, a new upgrade of 423 series. It is equipped with an Intel Core i3 CPU and up to 32 GB DDR5 memory, catering to the needs of individuals and business users who demand higher performance. The 424 series will come with TOS 5.1 operating system and can be smoothly upgraded to the TerraMaster TOS 6 in the future.

Peak Performance 4-Bay NAS
Intel Core i3 8-core 8-thread CPU at 3.8 GHz (turbo), integrated UHD GPU at 1.25 GHz, 32 GB DDR5, dual 2.5G Ethernet ports, and dual M.2 NVMe slots for SSD caching.

Chinese company Sipeed has introduced the Lichee Pocket 4A, one of the first handheld gaming devices based on the RISC-V open-source instruction set architecture (ISA). Sipeed positions the device as a retro gaming platform capable of running simple titles via software rendering or GPU acceleration. At its core is Alibaba's T-Head TH1520 processor featuring four 2.50 GHz Xuantie C910 RISC-V general-purpose CPU cores and an unnamed Imagination GPU. The chip was originally aimed at laptop designs. Memory options include 8 GB or 16 GB LPDDR4X RAM and 32 GB or 128 GB of storage. The Lichee Pocket 4A has a 7-inch 1280x800 LCD touchscreen, Wi-Fi/Bluetooth connectivity, and an array of wired ports like USB and Ethernet. It weighs under 500 grams. The device can run Android or Linux distributions like Debian, Ubuntu, and others.

As an early RISC-V gaming entrant, performance expectations should be modest—the focus is retro gaming and small indie titles, not modern AAA games. Specific gaming capabilities remain to be fully tested. However, the release helps showcase RISC-V's potential for consumer electronics and competitive positioning against proprietary ISAs like ARM. Pricing is still undefined, but another Sipeed handheld console retails for around $250 currently. Reception from enthusiasts and developers will demonstrate whether there's a viable market for RISC-V gaming devices. Success could encourage additional hardware experimentation efforts across emerging open architectures. With a 6000 mAh battery, battery life should be decent. Other specifications can be seen in the table below, and the pre-order link is here.

NETGEAR, Inc., the leading provider of networking products that power businesses both large and small, today launched the cloud-managed NETGEAR S3600 Series Smart Switches, joining the company's cutting-edge portfolio of switches designed to meet the growing demands for faster networking that today's businesses require. The S3600 Series includes the 8-Port 10G/Multi-Gigabit (XS508TM) and 16-Port 10G/Multi-Gigabit (XS516TM) that provide flexible and auto-adaptive wired connectivity of high-bandwidth devices.

The growing networking demands of high-bandwidth applications and devices in the workplace often result in connectivity drops, slow speeds, bottlenecks, lagging content streaming and slow downloads. The S3600 Series of Smart Switches addresses these issues head-on and enables business owners to simply connect devices and instantly give each device exactly the speed it needs - either 1G, 2.5G, 5G, or 10G.

Generative AI (GenAI) is the decade's most promising accelerator for innovation with 78% of IT decision makers reporting they're largely excited for the potential GenAI can have on their organizations.¹ Most see GenAI as a means to provide productivity gains, streamline processes and achieve cost savings. Harnessing this technology is critical to ensure organizations can compete in this new digital era.

Dell Technologies and AMD are coming together to unveil an expansion to the Dell Generative AI Solutions portfolio, continuing the work of accelerating advanced workloads and offering businesses more choice to continue their unique GenAI journeys. This new technology highlights a pivotal role played by open ecosystems and silicon diversity in empowering customers with simple, trusted and tailored solutions to bring AI to their data.

Marvell Technology, a leader in data infrastructure semiconductor solutions, is enabling networking equipment and firewall manufacturers achieve breakthrough levels of performance and efficiency with two new OCTEON 10 data processing units (DPUs), the OCTEON 10 CN102 and OCTEON 10 CN103. The 5 nm OCTEON CN102 and CN103, broadly available to OEMs for product design and pilot production, are optimized for data and control plane applications in routers, firewalls, 5G small cells, SD-WAN appliances, and control plane applications in top-of-rack switches and line card controllers. Several of the world's largest networking equipment manufacturers have already incorporated the OCTEON 10 CN102 into a number of product designs.

Containing up to eight Arm Neoverse N2 cores, OCTEON 10 CN102 and CN103 deliver 3x the performance of Marvell current DPU solutions for devices while reducing power consumption by 50% to 25 W. Achieving SPEC CPU (2017) integer rate (SPECint) scores of 36.5, OCTEON 10 CN102 and CN103 are able to deliver nearly 1.5 SPECint points per Watt. The chips can serve as an offload DPU for host processors or as the primary processor in devices; advanced performance per watt also enables OEMs to design fanless systems to simplify systems and further reduce cost, maintenance and power consumption.

Artificial intelligence has been a hot topic this year, and everything is now an AI processor, from CPUs to GPUs, NPUs, and many others. However, it was only a matter of time before we saw an integration of AI processing elements into the networking chips. Today, Broadcom announced its new Ethernet switching silicon called Trident 5-X12. The Trident 5-X12 delivers 16 Tb/s of bandwidth, double that of the previous Trident generation while adding support for fast 800G ports for connection to Tomahawk 5 spine switch chips. The 5-X12 is software-upgradable and optimized for dense 1RU top-of-rack designs, enabling configurations with up to 48x200G downstream server ports and 8x800G upstream fabric ports. The 800G support is added using 100G-PAM4 SerDes, which enables up to 4 m DAC and linear optics.

However, this is not only a switch chip on its own. Broadcom has added AI processing elements in an inference engine called NetGNT (Networking General-purpose Neural-network Traffic-analyzer). It can detect common traffic patterns and optimize data movement across the chip. Specifically, the company has listed an example of the system doing AI/ML workloads. In that case, NetGNT performs intelligent traffic analysis to avoid network congestion in these workloads. For example, it can detect the so-called "incast" patterns in real-time, where many flows converge simultaneously on the same port. By recognizing the start of incast early, NetGNT can invoke hardware-based congestion control techniques to prevent performance degradation without added latency.

Belkin, a leading consumer electronics brand for 40 years, today announced a new 11-in-1 dock featuring the latest Silicon Motion technology to add more power and productivity to any desk setup. This universal dock offers additional key ports that newer laptop models no longer support and can expand to three monitors with resolution of up to 4K at 60 Hz for one monitor.

With universal capabilities, devices stay charged with up to 100 W of pass-through power and fast transfer speeds of up to 10 Gbps. This dock also doubles as a laptop stand and includes a 350 mm centered connection cable for easy connection to a laptop in any space.

During its Ignite conference, Microsoft introduced a duo of custom-designed silicon made to accelerate AI and excel in cloud workloads. First of the two is Microsoft's Azure Cobalt 100 CPU, a 128-core design that features a 64-bit Armv9 instruction set, implemented in a cloud-native design that is set to become a part of Microsoft's offerings. While there aren't many details regarding the configuration, the company claims that the performance target is up to 40% when compared to the current generation of Arm servers running on Azure cloud. The SoC has used Arm's Neoverse CSS platform customized for Microsoft, with presumably Arm Neoverse N2 cores.

The next and hottest topic in the server space is AI acceleration, which is needed for running today's large language models. Microsoft hosts OpenAI's ChatGPT, Microsoft's Copilot, and many other AI services. To help make them run as fast as possible, Microsoft's project Athena now has the name of Maia 100 AI accelerator, which is manufactured on TSMC's 5 nm process. It features 105 billion transistors and supports various MX data formats, even those smaller than 8-bit bit, for maximum performance. Currently tested on GPT 3.5 Turbo, we have yet to see performance figures and comparisons with competing hardware from NVIDIA, like H100/H200 and AMD, with MI300X. The Maia 100 has an aggregate bandwidth of 4.8 Terabits per accelerator, which uses a custom Ethernet-based networking protocol for scaling. These chips are expected to appear in Microsoft data centers early next year, and we hope to get some performance numbers soon.

Ayar Labs, a leader in silicon photonics for chip-to-chip connectivity, will showcase its in-package optical I/O solution integrated with Intel's industry-leading Agilex Field-Programmable Gate Array (FPGA) technology. In demonstrating 5x current industry bandwidth at 5x lower power and 20x lower latency, the optical FPGA - packaged in a common PCIe card form factor - has the potential to transform the high performance computing (HPC) landscape for data-intensive workloads such as generative artificial intelligence (AI), machine learning, and support novel new disaggregated compute and memory architectures and more.

"We're on the cusp of a new era in high performance computing as optical I/O becomes a 'must have' building block for meeting the exponentially growing, data-intensive demands of emerging technologies like generative AI," said Charles Wuischpard, CEO of Ayar Labs. "Showcasing the integration of Ayar Labs' silicon photonics and Intel's cutting-edge FPGA technology at Supercomputing is a concrete demonstration that optical I/O has the maturity and manufacturability needed to meet these critical demands."

NETGEAR, Inc., the leading provider of networking products that power businesses both large and small, today launched the NETGEAR MS108TUP 8-Port Multi-Gigabit (2.5G) Ethernet Ultra60 PoE++Smart Switch, purpose-built for growing businesses that need all-in-one, reliable, seamless connectivity.

The growing number of connected devices, use of data-and-power-hungry applications and increasing prevalence of "bring your own device" make it increasingly important for small and mid-sized businesses to have a modern networking solution to drive better performance. NETGEAR's cutting-edge line of Smart Switches is now joined by a new model that provides an option to business owners with smaller port count needs without sacrificing quality.