Good April day to you! It was a wild week for more than the HPC-AI sector last week, here’s a brief (7:39) look at some key developments: U.S. tariffs, the technology sector and advanced chips, Intel-TSMC ….
HPC News Bytes 20250407: Tariffs and the Technology Industry, Intel-TSMC Deal?, DARPA Taps Wafer Scale, Sandia and Laser-Based Cooling, Optical I/O News
April 7, 2025 by
Optical I/O: Ayar Labs Launches 8 Tbps UCIe Optical Chiplet for AI Scale-Up Architectures
April 1, 2025 by
Optical interconnect company Ayar Labs announced the TeraPHY optical I/O chiplet, which Ayar said is the first optical Universal Chiplet Interconnect Express (UCIe), designed to maximize AI infrastructure performance and efficiency while reducing latency and power consumption.
@HPCpodcast: With Hyperion Research on the Big Trends Across HPC-AI, Quantum, Cloud, Optical, Storage and Interconnects
February 28, 2025 by
We dive into numbers coming out Hyperion’s latest HPC-AI market update, including the strong current and projected industry CAGRs, the state of quantum computing (including thoughts …
@HPCpodcast: Addison Snell on the Present and Future of HPC-AI, Quantum, DeepSeek, Interconnects, Leadership Supercomputing
February 7, 2025 by
We delve into current market sizing for HPC-AI (the conjoining of the two being a relatively new focus), a look ahead at the rest of the year and the decade, the impact of hyperscalers on AI ….
Emerald Invests in NLM Photonics Data Center Efficiency
January 29, 2025 by
SEATTLE, January 29, 2025 – Emerald, a venture capital company, has invested in NLM Photonics to help bring energy-efficient, high performance electro-optic modulation technology to AI, data centers, quantum computing, and more. Other investors include Oregon Venture Fund, Tokyo Ohka Kogyo Co., Ltd., Pack Ventures, StoryHouse Ventures, and other existing NLM investors to support NLM’s […]
Photonic Interconnect: Lightmatter Joins UALink Consortium
December 18, 2024 by
Dec. 18, 2024, MOUNTAIN VIEW, Calif. –Photonic interconnect company Lightmatter has joined the Ultra Accelerator Link Consortium (UALink), an industry initiative focused on creating high-speed, low-latency open interconnect standards for scale-up AI systems. As a Contributor member, Lightmatter will apply its Passage technology—a 3D-stacked photonic engine—for accelerator-to-accelerator communication, enabling unprecedented performance in AI clusters. The UALink […]
Optical Interconnect: Ayar Labs Secures Investments from AMD, Intel Capital, NVIDIA in $155M Round
December 11, 2024 by
SAN JOSE – Dec. 11, 2024 – Ayar Labs, an optical interconnect company for large-scale AI workloads, today announced it has secured $155 million in financing led by Advent Global Opportunities and Light Street Capital. This brings the company’s total funding to $370 million and raises the company’s valuation to above $1 billion. The strength of the […]
Lightmatter and Amkor Technology Partner on 3D Photonics Package
November 14, 2024 by
MOUNTAIN VIEW, Calif., Nov. 14, 2024 — Photonic supercomputing company Lightmatter today announced a strategic partnership with Amkor Technology, Inc. (Nasdaq: AMKR), a provider of semiconductor packaging and test services, to create the largest-ever 3D-packaged chip complex utilizing Lightmatter’s innovative Passage platform. This collaboration harnesses Lightmatter’s groundbreaking 3D-stacked photonic engine along with Amkor’s advanced multi-die packaging expertise to […]
Optical I/O: Celestial AI Acquires Rockley Photonics Portfolio
October 22, 2024 by
SANTA CLARA, Calif. – October 23, 2024 – Photonic fabric company Celestial AI today announced the acquisition of silicon photonics intellectual property from Rockley Photonics (Rockley), including worldwide issued and pending patents. Celestial said the combination results in one of the strongest IP portfolios in the field of silicon photonics for optical compute interconnect, bringing their […]
Optical I/O: The Key to Unlocking AI Infrastructure, Profitability and Performance
October 1, 2024 by
[SPONSORED GUEST ARTICLE] Scaling GenAI inference performance requires increasing the number of GPUs or accelerators working in parallel within the scale-up domain. In-package optical I/O offers a path forward by breaking the ….
