SIMULIA’s Post

Startup Lightmatter Advances Light-Speed Computing Meet Lightmatter, the #startup poised to redefine the very essence of computing as we know it. Ditching traditional #electricity in favor of harnessing the power of light. Lightmatter embarks on a mission to revolutionize how information is processed and improve on the work by for example Ayar Labs Founded in 2017 by Nicholas Harris, Darius Bunandar, and Thomas Graham, pioneers light-based computing for faster, energy-efficient processing, stemming from #MIT. By utilizing light-based processors known as photonic chips, Lightmatter's computers can perform complex tasks with unprecedented efficiency. These chips manipulate light waves to perform calculations, offering a promising alternative to traditional silicon-based processors. Lightmatter's unique approach is its ability to perform multiple computations simultaneously, a feat known as parallel processing. This allows for faster and more efficient data processing, making Lightmatter's technology particularly well-suited for tasks such as #artificialintelligence and data analytics. Lightmatter's technology also prioritizes energy efficiency, standing in contrast to energy-intensive computing giants like OpenAI and NVIDIA. By reducing computing energy consumption, could Lightmatter pave the way for a more #sustainable digital future?

🔓 Unlocking New Possibilities: The Future of Electronic Design 🔓 At OpenQQuantify our AI-driven models are more than just tools—they’re a revolution in how electronic design is conceived and executed. We’re transforming the landscape, enabling businesses and innovators to push the boundaries of what’s possible in technology. Whether you’re looking to enhance your product development process, accelerate innovation, or stay ahead in a competitive market, our cutting-edge solutions are designed to help you achieve your goals with unprecedented efficiency and precision. Don’t just adapt to the future—shape it. Explore the possibilities with OpenQQuantify https://lnkd.in/ezFz8_9t #AI #Electronics #Innovation #Technology #FutureDesign #Engineering #Revolution

#CXL Is Dead In The #AI Era (by SemiAnalysis) AI Accelerator Beachfront Considerations, Memory Pooling Negatives, Custom Silicon Adoption If we travel back 2 years in time, before the rapid rise in AI, much of the #datacenter #hardware world was chasing CXL. It was promised as the messiah to bring heterogenous compute, #memory pooling, and composable #server architectures. Existing players and a whole new host of #startups were rushing to integrate CXL into their products, or create new CXL based products such as memory expanders, poolers, and switches. Fast forward to 2023 and early 2024, and many projects have been quietly shelved and many of the #hyperscalers and large #semiconductor companies have almost entirely pivoted away. With the upcoming Astera Labs IPO and product launches, CXL discussions are coming back to the front row at least for a short while. SemiAnalysis have written extensively about the #technology, its cost savings potential for #CloudService Providers, and the ecosystem & hardware stack. While very promising on paper, the datacenter landscape has evolved a lot, but one thing hasn’t changed: CXL hardware such as controllers and switches are still not shipping in meaningful volumes. Despite this, there is still a lot of noise and research around CXL, with certain professionals in the industry now pushing the narrative of CXL as an 'enabler' of AI. Is the broader CXL market ready to take off and deliver on its promise? Can CXL become the interconnect for AI applications? What is the role in CPU attach expansion and pooling? SemiAnalysis will answer these questions in the subscriber section of their latest report via the link below 💡🙏👇 https://lnkd.in/eprYxgdE The simple answer is no - the people pushing CXL for AI are dead wrong. Let’s start with a quick reminder of the main use cases and promises of CXL. A Big thank you again to Dylan Patel (sorry could not see you this time in #Taiwan - but next time again! , Jeremie Eliahou Ontiveros and SemiAnalysis #semiconductorindustry #semiconductors #semiconductormanufacturing #technology #tech #it #hpc #dram #computing #computer #chip #chips #ic

𝗨𝗻𝗰𝗼𝘃𝗲𝗿𝗶𝗻𝗴 𝗖𝗼𝗼𝗹𝗶𝗻𝗴 𝗖𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲𝘀 𝗶𝗻 𝗛𝗶𝗴𝗵-𝗗𝗲𝗻𝘀𝗶𝘁𝘆 𝗗𝗮𝘁𝗮 𝗖𝗲𝗻𝘁𝗿𝗲𝘀 ❄️💻 In part one of a sponsored interview with Techerati, Chatsworth Products' Samuel Rodriguez, discussed the critical thermal management challenges facing modern data centres. As AI, high-performance computing (HPC), and high-density racks become mainstream, Sam emphasised the need for innovative #cooling solutions tailored to evolving demands. 𝗧𝗵𝗲 𝗥𝗶𝘀𝗶𝗻𝗴 𝗖𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲 𝗼𝗳 𝗛𝗶𝗴𝗵-𝗗𝗲𝗻𝘀𝗶𝘁𝘆 𝗖𝗼𝗼𝗹𝗶𝗻𝗴 📈🌡️ Over the past decade, data centres operated with air-cooled designs at lower kilowatt ranges per rack. However, the rise of advanced applications like ChatGPT and high-performance GPUs has shifted power requirements, with processors like NVIDIA’s H100X SXM now reaching up to 700 watts. "AI has pushed thermal management into a critical area," Sam explained, highlighting the increased complexity in managing higher #power densities effectively. Read more at https://lnkd.in/e88EH72z #technews #thermalmanagement #datacenters #coolingsolutions #liquidcooling #ai #chatgpt

Meet Spacetop G1, the world's first screenless laptop. A Tech Startup known as "Sightful" has just unveiled its contribution to the spacial computing space. Spacetop G1, also known as the "AR laptop for work" uses connected glasses that projects a 100-inch screen that can either be used to open a single monitor or split into multiple workspaces. The CEO of Sightful, Tamir Berliner has this to say about this development: “We introduced Spacetop to free people from the constraints of time and space, as we don’t believe people should be stuck in 14” screens,” ------------------------------------------------------------------------------ Would you get one? Follow Obaloluwa for updates on everything AI and Tech.

AI and Power Needed WSJ reported "Tech companies scouring the country for electricity to power artificial intelligence are increasingly finding there is a waiting list. In many places the nation’s high-voltage electric wires are running out of room, their connection points locked up by data centers for AI, new factories or charging infrastructure for electric vehicles.  A mad dash to lock up available power has ensued.  The tech industry is pinballing from one market to the next looking for places with the capacity to connect campuses that would consume up to a gigawatt of power—about as much as San Francisco uses. Some requests are as much as four to five times as large as that. But wires are getting so crowded that some prospective data center customers—which request far more power than other users—are being told they may have to wait until the next decade to get the power they are seeking. Others are receiving less power than they expected."

#CXL cost, complexity in integration and utility vis-a-vis #RAM was always a question mark for scale users, now it's largest customers, #datacentres and #cloud system have parked it outside to find better grounds for it in the unforeseen future.. #CXL can still look for #architectural pivots that maybe favoring them at #edge.. starting small with large volumes many times never hurts.. #isCXLdead? Thanks Marco Mezger for sharing this..

In the world of technology, few innovations have the power to transform entire industries as profoundly as computer vision. This technology, which allows computers and systems to see and interpret the environment around them, is at the forefront of the fourth industrial revolution. 🖥️ Computer vision uses advanced algorithms to process and analyze images and videos, enabling machines to perform complex tasks, from identifying faults and irregularities in production lines to automated monitoring of public spaces for security. 🌐 The applicability of this technology extends across various sectors, including healthcare, retail, manufacturing, and many other fields. The possibilities are endless, and the benefits for businesses and society as a whole are immeasurable. What challenges do you see that could be solved with this technology? Share your ideas and let's innovate together. #ComputerVision #Innovation #Technology #Industry4.0 #DigitalTransformation

Startups are the lifeblood of our industry, and we always check the EE Times | Electronic Engineering Times Silicon 100 list each year to see where the action is. Now in its 24th edition and compiled since its inception by eminent journalist Peter Clarke, the Silicon 100 provides a fascinating analysis of what’s going on around the world in the electronics sector, broken down by criteria such as technology and location. This year, it’s perhaps no surprise to see increased representation of companies working on #AI and #chiplets, two of the most hyped areas right now. However, there are plenty of other innovations making their mark, from #quantumcomputing to #MEMS to #5G. Geographically, California remains the center of the startup world, but there’s a steadily increasing European presence on the list. So which of the 100 companies caught our eye? There are too many interesting firms to mention, but a few highlights are Finland’s IQM Quantum Computers, and the UK’s Paragraf®, which has developed a process for producing single-atomic-layer materials on substrates such as silicon and GaN. And while there are multiple companies on the list working on #edgeAI (a technology we’ve been involved with for several years), one that stands out is Swiss firm Synthara AG, with its innovations for in-memory computing. Read the list here: https://lnkd.in/gwuSJ2fQ And let us know in the comments which startups on the list caught your attention. #innovation #technology #sustainability #electronics #semiconductorindustry