MRC Group’s Post

#ICYMI - Last week we took advantage of being a #MemCon founding partner to showcase some of our latest technologies at #MemCon2024. One such technology is our #HBM3E 12H, the industry’s first developed 12-stack HBM3E DRAM. Designed to meet the memory bandwidth requirements of next-gen AI applications, HBM3E 12H offers bandwidth of up to 1,280 GB/s and capacity up to 36GB, a more than 50% improvement compared to its predecessor that dramatically reduces the total cost of ownership (TCO) to data centers. Additionally, attendees could view our CXL Memory Module – DRAM (CMM-D) technology, which uses DRAM integrated with the CXL open standard interface. We also had a number of our experts along with some partners present on CXL and HBM innovations and solutions. Overall, it was an exceptional event for #SamsungMemory! Check out moments from #MemCon2024, all our featured technologies, and more to see how we are building a memory ecosystem to overcome the “memory wall” together and accelerate a hyperscale AI era: https://lnkd.in/gWazQRnn #innovation #memory #CXL #AI #SamsungSemiUS

PowerSwitch Ethernet fabric provided zero-loss handling of AI traffic from the 64 connected NICs/GPUs. Read the full report here: https://lnkd.in/dkHKPDxg #iwork4dell

🚀 Dive into the future of computing with Canalys’ latest report on AI-capable PCs! 🤖💻 Unveiling insights into the revolutionary impact of AI on personal computing in 2024 and beyond. In this 20-page report, we explore how AI is reshaping the tech landscape and creating opportunities for the future. We anticipate a whopping 60% of PCs shipped by 2027 to be AI-capable, signaling a significant technological transformation. Discover the role of specialized AI chipsets in enhancing productivity, leisure experiences, and more. Get the full complimentary report of the Now and Next in AI PCs: https://lnkd.in/gWgYHjiU #GenAI #AIPC #PC #aichips #artificialintelligence #NextGen

TechCrunch: ‘SiMa.ai secures $70M funding to introduce a multimodal GenAI chip’ “As the demand for GenAI is growing, SiMa.ai is set to introduce its second-generation ML SoC in the first quarter of 2025 with an emphasis on providing its customers with multimodal GenAI capability. The new SoC will be an ‘evolutionary change’ over its predecessor with ‘a few architectural tunings’ over the existing ML chipset, CEO Krishna Rangasayee said. The new GenAI SoC would adapt to any framework, network, model and sensor — similar to the company’s existing ML platform — and will also be compatible with any modality, including audio, speech, text and image.” Jagmeet Singh of TechCrunch provides a deep dive into this week’s funding news and SiMa.ai’s vision for ONE platform. Read more below! #EdgeComputing #AI #ML #GenAI

Can FPGAs provide flexible computing for #Edge AI? The world of AI typically relies on multi-cluster GPUs and other dedicated hardware accelerators. FPGA-based #Edge AI and #vision technology are uncommon, and it seems that Mirametrix's Glance is making a significant impact in this area. I believe that configurable #smart edge devices, whether used for camera (#objectdetection) or for providing #real-time local contextual feedback, would greatly benefit from this breakthrough.

📈 Explosive Growth in Data Communication Optical Components! Cignal AI's latest 4Q23 Optical Components Report reveals a booming demand for optical products driven by AI cluster applications. For the fourth consecutive quarter, data communication optical component revenue has surged. NVIDIA, Coherent, and InnoLight lead the way in 800GbE module shipments for hyperscale AI applications. Acacia and Marvell dominate in high-performance coherent interface shipments, thanks to the widespread adoption of 400ZR pluggable modules. Scott Wilkinson, Cignal AI’s Lead Optical Component Analyst, states, “The shipment of data communication products—especially 800GbE optics—is rapidly increasing, with 2024 projections hitting 8 million units. While the telecom sector is slowly recovering after bottoming out in Q3 2023, a swift rebound is unlikely in the short term.” #OpticalComponents #DataCommunication #800GbE #AI #Telecom #IndustryTrends #CignalAI

Interoperability is one of the fundamental challenges facing AI’s use in wireless devices and networks. We’re excited to announce that Nokia Bell Labs and Qualcomm have developed a prototype technology that utilizes sequential learning to overcome this problem. By enabling multi-vendor interoperability between the different AI models used in networks and devices, future wireless systems would support the use of AI to achieve greater capacities and better performance while potentially reducing their energy consumption. Visit us at MWC to see a proof-of-concept, or read more here to find out how this technology will save time and help scalability for future real-world deployments: https://nokia.ly/49GTQMA #NokiaInnovation #Connectivity #AI #MWC24

The future of #AI will not be possible without continuous advances in optical interconnects, which are foundational technologies that lower both power needs and costs while also scaling AI infrastructure with high bandwidth connectivity. The good news is that Broadcom is leading the industry in core optical technologies for AI data centers, across VCSEL, EML and CPO. Learn more about this portfolio of solutions from Near Margalit, VP and GM of Broadcom’s Optical Systems Division: https://lnkd.in/grFS_rWm

The future of #AI will not be possible without continuous advances in optical interconnects, which are foundational technologies that lower both power needs and costs while also scaling AI infrastructure with high bandwidth connectivity. The good news is that Broadcom is leading the industry in core optical technologies for AI data centers, across VCSEL, EML and CPO. Learn more about this portfolio of solutions from Near Margalit, VP and GM of Broadcom’s Optical Systems Division: https://lnkd.in/gBJXDVQm

The potential for chiplet technology to be a transformational paradigm is now widely recognized. The cost, time-to-market, and power consumption benefits of chiplet-based solutions are compelling the industry toward integrating multiple dies in a single package. AI has emerged as a primary catalyst for this trend. Custom silicon designed for AI benefits significantly from the chiplet approach, which combines dense logic and memory with the need for high-speed connectivity. The push for custom AI hardware is rapidly evolving, with a focus on energy-efficient designs. Chiplets offer the flexibility to create systems-in-package that balance cost, power, and performance for specific workloads without starting from scratch. AI's unique needs for inter-die communication make reducing latency crucial, and the rollout of larger clusters emphasizes the role of high-speed, optical interconnects. Join Tony Chan Carusone, CTO of Alphawave Semi in his session 𝗥𝗲𝗱𝗲𝗳𝗶𝗻𝗶𝗻𝗴 𝗖𝗼𝗻𝗻𝗲𝗰𝘁𝗶𝘃𝗶𝘁𝘆: 𝗖𝗵𝗮𝗿𝘁𝗶𝗻𝗴 𝗡𝗲𝘅𝘁-𝗚𝗲𝗻 𝗣𝗮𝘁𝗵𝘄𝗮𝘆𝘀 𝗶𝗻 𝗖𝗵𝗶𝗽𝗹𝗲𝘁 𝗜𝗻𝘁𝗲𝗿𝗰𝗼𝗻𝗻𝗲𝗰𝘁𝘀 Meet Tony and 80 other global speakers at the AI Hardware & Edge AI Summit, San Jose, CA, on 9-12th September 2024 at Signia by Hilton, San Jose, CA. Learn more: https://lnkd.in/eWfkuckF #AIHWEdgeAISummit #AI #EdgeAI #AIHardware