Andy Patrizio is a freelance technology writer based in Orange County, California. He's written for a variety of publications, ranging from Tom's Guide to Wired to Dr. Dobbs Journal, and has been on staff at IT publications like InternetNews, PC Week and InformationWeek.
Nvidia's Grace superchips cater to AI as well as legacy high-bandwidth applications not optimized for GPUs.
Based on a new ASIC, DPU, smartNIC, and SDK, Nvidia ups the speed and efficiency of its switching platform.
Nvidia’s H100 GPU is the first in its new family of Hopper processors that the company claims will underpin the world’s fastest supercomputer.
Memory is performance, and DDR5 promises much faster memory at a lower power draw.
Marvell announced a new Arm-based smartNIC, and Dell will be the first to use it.
As supply-chain woes make Wi-Fi 6E gear hard to find, enterprises are opting to delay upgrades until Wi-Fi 7 is available late next year.
New services from Verizon are specifically designed around supporting edge networks.
PowerVault ME5 doubles the performance and capacity of Dell's ME4 storage platform and is optimized for SAN and DAS.
With lead times as long as 400 days, enterprises need to start looking at alternative ways to get the network equipment they need.
HPE Private 5G enables hybrid wireless networks that support both 5G and Wi-Fi to handle specific enterprise needs.
Intel's CEO lays out two tracks for its Xeon chip development through 2024 with implications for servers, supercomputing, and AI.
Intel is setting up an internal group to develop a processor optimized for Bitcoin mining, but the group will also work on "other custom accelerated-supercomputing opportunities at the edge".
AWS, Microsoft Azure, and Google Cloud dominate cloud services.
Arm will now be spun off with an IPO, expected to be completed before March 2023
Chip giant goes from not invented here attitude to a more open mindset.
More than 20 vulnerabilities have been found affecting unified extensible firmware interfaces (UEFI) software, allowing attackers to bypass hardware security mechanisms.
Infinidium's enclosed, floor-to-ceiling chamber sucks in cool air that draws heat from servers, rises, and drives a turbine as it exits the building through a vent at the top.
Organizations of all sizes are ready to move on from COVID-19 stopgap projects and initiate major IT investments, according to Gartner, which forecasts in surge in cloud spending and managed IT services.
AMD's $35 billion all-stock deal for Xilinx is now expected to close this quarter after more than a year of delays.
The Nvidia-Arm deal that would set up Nvidia as a more formidable processor powerhouse faces strong opposition from the UK, China.
The speed of Meta's Research Super Computer would dwarf that of the current world's fastest supercomputer.
7 new Dell systems and services cover multi-cloud offerings and more.
Chip giant teams with startup GRC to take immersion cooling mainstream.
The ThinkEdge SE450 incorporates Nvidia GPUs for enterprise and industrial AI at the edge.
Buying Bright Computing gives Nvidia in-house clustering software for high-performance computing and includes support for VMware.
Due to a “date issue,” mail was being queued up instead of sent.
Framework is designed to help data-center operators minimize the environmental impacts of their facilities.
Seven organizations are looking to develop a fuel cell platform for data centers.
ITRenew will brings its expertise in IT hardware reuse and disposal to Iron Mountain’s IT asset lifecycle management business.
Equinix Metal service now includes Nvidia, Intel, AMD, and Ampere processors.
Intel is working with partners in China to build an infrastructure processing a unit that relies on both a CPU and an FPGA.
US regulators are investigating whether allowing Nvidia to buy Arm would give it too much control over the semiconductor market and hurt competition.
While missing some details, the Arm-based processor has some key new performance metrics.
Customers may find their appliances compromised or a screw loose.
The Xilinx Alveo U55C data-center accelerator is smaller, has more memory, and draws less power than its predecessor, making it more attractive for high-performance computing.
The new Nvidia Bluefield-3 DPU will be used to provision next-gen firewall services from major providers.
Nvidia is blending its BlueField data-processing units with its Morpheus AI security framework to contribute to a Zero Trust architecture.
Powerful new server CPU and a GPU it says tops Nvidia’s best are part of the parade.
Pluribus software on ITRenew hardware means an open architecture for compute and storage infrastructure.
Billed as a drop-in replacement for Ethernet and InfiniBand cabling and switches, Rockport's switchless network relies on endpoint network cards.
NetApp continues to evolve its cloud strategy while maintaining on-premises support.
Run.AI is launching software that orchestrates shared use of organizations' GPU resources rather than dedicating entire processors to specific AI workloads.
New AI servers can boost performance into supercomputing territory with a smaller footprint.
Virtual instances of new designs will give developers a head start on creating apps before the silicon arrives.
With the Loihi 2 neuromorphic chip, machines can perform application processing, problem-solving, adaptation,and learning much faster than before.
The managed-infrastructure-services unit of IBM's Global Technology Services faces stiff competition when it goes solo as Kyndryl later this year.
The Octeon 10 DPU packs the processing power of a server CPU but it’s for network management.
If ever you needed a sign Dell and VMware are parting on good terms, they just gave it.
The company boasts of massive performance gains, but is anyone buying?
NAND flash maker's Software-Enabled Flash technology is designed to add customization to SSDs.
Sponsored Links