Tag
西班牙巴塞罗那超级计算中心（BSC）的AI Factory加速器计划正式投入运营，多家初创企业借助NVIDIA Inception计划与MareNostrum 5超算资源，在医疗、水务、制药等领域推出AI创新应用。
西班牙计算历史博物馆用30台2012年的Mac Mini组建集群，打造了一台1:1比例的Cray-1超级计算机复制品，实现约1.3 teraflops性能，作为Cray-1与苹果创立50周年的致敬项目。
MIT Lincoln Laboratory researchers have published a series of papers under the Lincoln AI Computing Survey (LAICS) documenting and comparing the performance and power of commercial AI accelerators — from a first study of 57 chips to a latest edition covering over 120.
A look inside the new Bull factory in Angers, which manufactures the biggest AI training clusters in Europe, noting the scale of the facility as impressive.
SpaceX's supercomputing facilities in the Mid-South rank among the world's most advanced AI training clusters, featuring millions of GPUs and over two gigawatts of compute, while the company is actively hiring.
The EU is building up to seven AI gigafactories with up to €10bn in public funding, aiming to boost European AI computing capacity and provide access for businesses of all sizes. The article examines which mid-sized companies will actually benefit and the timeline, with operations unlikely before 2028.
The Well is an open-source project providing 15TB of high-precision physics simulation results, which would normally cost millions of dollars and weeks on a supercomputer at a national lab, now 100% free and open source, making a huge impact on the scientific research field.
Chinese supercomputer LineShine has claimed the top spot in the global supercomputer rankings, marking a significant achievement in high-performance computing.
The LineShine supercomputer in Shenzhen, China, claims the number 1 spot on the TOP500 list with 2.198 Exaflops of sustained FP64 performance, powered by a custom Armv9 CPU with 13 million cores. It also leads the HPCG benchmark, surpassing El Capitan.
EU's Frontier AI Grand Challenge winner is the EUROPA consortium led by Domyn, which will build an open-source 400B+ parameter model covering 24 EU languages using European supercomputers; the prize is compute capacity, not cash.
NVIDIA technology now powers over 400 of the world's 500 fastest supercomputers (81% of the TOP500), with record GPU and networking adoption and top efficiency on the Green500 list.
JUPITER, Europe's first exascale supercomputer powered by NVIDIA Grace Hopper, demonstrated four groundbreaking projects at ISC: mapping the human brain at cellular scale, simulating climate at 1km resolution, building AI for wireless networks, and simulating a 50-qubit quantum computer.
NVIDIA announces its Vera CPU will power new supercomputers at Los Alamos National Laboratory, delivering significant performance improvements for agentic AI simulations and scientific workloads.
NVIDIA announces live updates from GTC Taipei at COMPUTEX, including COMPUTEX Best Choice Awards for Vera Rubin NVL72, Jetson Thor, and Alpamayo, showcasing breakthroughs in AI factories, robotics, and autonomous vehicles.
Thinking Machines Lab is hiring supercomputing engineers in NYC and SF to build infrastructure for real-time interactive models and large-scale training.
The article explores Armenia's strategic pivot toward AI and ICT as a means of digital reindustrialization following post-Soviet economic shifts. It details a major initiative involving Firebird, Dell, and NVIDIA to invest $500 million in local AI supercomputing infrastructure.
OpenAI has released MRC (Multipath Reliable Connection), a novel networking protocol developed with industry partners to improve performance and resilience in large-scale AI training clusters. The specification was published via the Open Compute Project to standardize infrastructure for efficient supercomputer operations.
Researchers from University College London published findings in Science Advances showing that a quantum computer connected to a supercomputer at the Leibniz Supercomputing Centre can enhance AI model predictions for complex fluid dynamics simulations, demonstrating practical quantum advantage over classical computing alone.
OpenAI submitted a response to the Department of Energy regarding AI infrastructure development on federal lands, advocating for streamlined permitting, predictable lease agreements, and financial incentives to accelerate deployment of AI supercomputer hubs while highlighting their partnership with Los Alamos National Laboratory.
An OpenAI backend engineer shares their personal journey into programming and describes their work maintaining and optimizing OpenAI's large-scale supercomputing clusters used for AI model training. The post highlights the complexity and scale of infrastructure challenges encountered at OpenAI.