Potential of Wireless Interconnects For Improving Performance And Flexibility Of Multi-Chip AI Accelerators


A new technical paper titled "Exploring the Potential of Wireless-enabled Multi-Chip AI Accelerators" was published by researchers at Universitat Politecnica de Catalunya. Abstract "The insatiable appetite of Artificial Intelligence (AI) workloads for computing power is pushing the industry to develop faster and more efficient accelerators. The rigidity of custom hardware, however, conflict... » read more

Co-Packaged Optics To Train/Run GenAI Models in Data Centers (IBM)


A new technical paper titled "Next generation Co-Packaged Optics Technology to Train & Run Generative AI Models in Data Centers and Other Computing Applications" was published by researchers at IBM. Abstract "We report on the successful design and fabrication of optical modules using a 50 micron pitch polymer waveguide interface, integrated for low loss, high density optical data transf... » read more

Two Silicon Switches Driven Both Thermally and Electrically With Ultra-Low-Crosstalk (Cambridge)


A new technical paper titled "Ultra-low-crosstalk Silicon Switches Driven Thermally and Electrically" was published by researchers at University of Cambridge and GlitterinTech Limited. Abstract "Silicon photonic switches are widely considered as a cost-effective solution for addressing the ever-growing data traffic in datacenter networks, as they offer unique advantages such as low power co... » read more

CXL-Based Heterogeneous Systems: How to Optimize and Future Directions (UCSD, Samsung, SK Hynix)


A new technical paper titled "The Hitchhiker’s Guide to Programming and Optimizing CXL-Based Heterogeneous Systems" was published by researchers at UC San Diego, Samsung, SK hynix. Abstract "We present a thorough analysis of the use of CXL-based heterogeneous systems. We built a cluster of server systems that combines different vendor's CPUs and various types of CXL devices. We further ... » read more

Visualization of Photoexcited Charges Moving Across the Interface of Si/Ge


A technical paper titled "Imaging hot photocarrier transfer across a semiconductor heterojunction with ultrafast electron microscopy" was published by researchers at UC Santa Barbara and UCLA. "In this work, we apply scanning ultrafast electron microscopy to provide a holistic view of photoexcited charge dynamics in a Si/Ge heterojunction. We find that the built-in potential and the band off... » read more

Workload-Specific Data Movements Across AI Workloads in Multi-Chiplet AI Accelerators


A new technical paper titled "Communication Characterization of AI Workloads for Large-scale Multi-chiplet Accelerators" was published by researchers at Universitat Politecnica de Catalunya. Abstract "Next-generation artificial intelligence (AI) workloads are posing challenges of scalability and robustness in terms of execution time due to their intrinsic evolving data-intensive characteris... » read more

Method To Determine The Permittivity of Dielectric Materials in 3D Integrated Structures At Broadband RF Frequencies


A new technical paper titled "Characterizing the Broadband RF Permittivity of 3D-Integrated Layers in a Glass Wafer Stack from 100 MHz to 30 GHz" was published by researchers at NIST. Abstract "We present a method for accurately determining the permittivity of dielectric materials in 3D integrated structures at broadband RF frequencies. With applications of microwave and millimeter-wave ele... » read more

Flip-Chip Bonding Technique To Excite LN Resonators Via Noncontact Electrodes (Yale)


A new technical paper titled "Noncontact excitation of multi-GHz lithium niobate electromechanical resonators" was published by researchers at Yale University. Abstract "The demand for high-performance electromechanical resonators is ever-growing across diverse applications, ranging from sensing and time-keeping to advanced communication devices. Among the electromechanical materials being ... » read more

DL Compiler for Efficiently Utilizing Inter-Core Connected AI Chips (UIUC, Microsoft)


A new technical paper titled "Scaling Deep Learning Computation over the Inter-Core Connected Intelligence Processor" was published by researchers at UIUC and Microsoft Research. Abstract "As AI chips incorporate numerous parallelized cores to scale deep learning (DL) computing, inter-core communication is enabled recently by employing high-bandwidth and low-latency interconnect links on th... » read more

Characterizing Three Supercomputers: Multi-GPU Interconnect Performance


A new technical paper titled "Exploring GPU-to-GPU Communication: Insights into Supercomputer Interconnects" was published by researchers at Sapienza University of Rome, University of Trento, Vrije Universiteit Amsterdam, ETH Zurich, CINECA, University of Antwerp, IBM Research Europe, HPE Cray, and NVIDIA. Abstract "Multi-GPU nodes are increasingly common in the rapidly evolving landscape... » read more

← Older posts Newer posts →