DCD Magazine #62 - The coming wave
Will compute at sea sink or swim?
News - Software, Charlotte Trueman, DatacenterDynamics
The latest news from the AI, data center, telco, chip, and cloud sector
DCD Magazine #62 out now
-
Riot Platforms agrees 191MW, 20-year lease agreement with Anthropic worth $9.1bn – report
Capacity will be deployed at Bitcoin miner’s campus in Rockdale, Texas
-
Anthropic publicly confirms it is putting together an in-house chip design team
Rumors AI company was looking to develop its own custom silicon first emerged earlier this year
-
NSF plans $100m AI Infrastructure Hubs program to boost science R&D
Initiative aims to address lack of AI infrastructure access in some parts of the US
-
Velox, I/ONX HPC to deploy 7,000 FuriosaAI RNGD accelerators at Swedish data center
Data center expected to have a total eventual capacity of 15MW
-
Nvidia and SK Group announce $500bn AI agreement, includes 2GW of data center capacity
First AI factory slated to come online in 2027 alongside broader memory partnership
-
DeepSeek to develop its own custom AI chip – report
Inference chip will allow company reduce its reliance on Nvidia and Huawei
-
Nvidia pushes Kyber release to 2028 following manufacturing concerns – report
Research firm SemiAnalysis claims 12 months delay is due to issues manufacturing the PCB midplane
-
Dell launches PowerRack, a turnkey compute, storage, and networking solution, updates Nvidia AI Factory platform
Company says it is working to close the gap between AI ambition and outcomes
-
Google unveils eighth-generation TPUs, two dedicated training and inference chips
“Purpose-build architectures” designed to support model training and inference workloads
-
Nvidia launches open source Ising AI models to support quantum chip development
The Ising calibration and decoding tools aim to tackle quantum error correction
-
Arm partners with Meta for data center AGI CPU
Marks the first time Arm has produced a physical chip, rather than licensing the IP
-
Synopsys launches AI-powered EDA solutions, including first integrated Ansys tools since acquisition
Announcements made at company’s annual Converge conference in California
-
IBM unveils new FlashSystem offerings, can be co-managed by its AI agents, says company
Product portfolio will be available from March 6
-
Tata Consultancy Services to offer AMD CPUs and GPUs as part of collaboration to boost AI adoption
Companies will jointly develop “industry specific” AI and generative AI solutions to support applications
-
Salute launches direct-to-chip liquid cooling service for AI and HPC facilities
Company said it is already working with a number of data center companies to support their DTC operations
-
ChipAgents raises $21m to support development of AI chip design platform
Has added former Siemens, Synopsys, and Cadence execs to its advisory board
-
Groq and IBM partner to boost delivery of agentic AI capabilities
IBM customers can now access GroqCloud via IBM watsonx Orchestrate
-
Meta partners with Arm for efficient AI system scaling on chip designer’s Neoverse platform
Companies also working to optimize Meta’s AI infrastructure software stack
-
IBM Spyre AI accelerator available later this month
Used to power IBM z17, LinuxONE 5 systems, and Power11 servers
-
Qualcomm acquires open-source hardware and software company Arduino
Follows acquisitions of Edge Impulse and Foundries.io
-
AMD signs agreement with generative AI startup Cohere for expanded use of Instinct GPUs
Comes as Cohere reveals $7bn valuation off the back of latest funding round
-
OpenAI to start mass-producing its custom AI chip in 2026 - report
Chip has been developed in partnership with Broadcom
-
Macquarie and Dell to bring Nvidia-powered sovereign AI to Australia
Companies will deploy a Dell ‘AI Factory’ at Macquarie’s IC3 Super West data center in Sydney
-
EnCharge AI launches its analog in-memory EN100 AI accelerator
Chip is available in two form factors for laptops, workstations, and Edge devices
-
Nvidia H200 outperforms Intel Gaudi 3 by factor of nine across first Llama 3.1 405B benchmark test – exclusive
Test was carried out by data center operator Aquatron in partnership with the Los Alamos National Laboratory
-
EdgeRunner AI raises $17.5m to develop air-gapped AI
Has already partnered with US Air Force
-
Groq secures $1.5bn from Saudi Arabia to expand AI inference infrastructure in the region
Investment will support expansion of Dammam data center
-
Alibaba, Apple, and Synopsys become latest members of UALink Consortium
Consortium now has more than 65 members