Chip Industry Technical Paper Roundup: May 26


New technical papers recently added to Semiconductor Engineering’s library: Technical Paper Research Organizations SHIP: SRAM-Based Huge Inference Pipelines for Fast LLM Serving 🔗 Nvidia, Groq Not All Thoughts Need HBM: Semantics-Aware Memory Hierarchy for LLM Reasoning 🔗 USC, University of Wisconsin-Madison Water-based, large-scale transfer of... » read more

Evaluating and Calibrating Performance On RISC-V Vector Processors (KTH, LLNL, BSC)


A new technical paper, "Closer in the Gap: Towards Portable Performance on RISC-V Vector Processors," was published by researchers at KTH Royal Institute of Technology, Lawrence Livermore National Laboratory, and Barcelona Supercomputing Center. Abstract "The RISC-V Vector Extension~(RVV) is a cornerstone for supporting compute throughout in scientific and machine learning workloads. Yet ... » read more

Chip Industry Technical Paper Roundup: May 11


New technical papers recently added to Semiconductor Engineering’s library: Technical Paper Research Organizations Source-position-dependent transmission cross coefficient formula including polarization and mask three-dimensional effects in High NA EUV🔗 Science Tokyo Performance and Energy Benefits of MRDIMMs 🔗 Barcelona Supercomputing Center, UPC, ... » read more

A Detailed Evaluation of A Production Server With High-End MRDIMM Main Memory (BSC, Micron, Intel, UPC)


A new technical paper, "Performance and Energy Benefits of MRDIMMs," was published by researchers at Barcelona Supercomputing Center, Universitat Politecnica de Catalunya, Micron and Intel Corporation. Abstract "Multiplexed Rank DIMMs (MRDIMMs) have recently emerged as memory devices that enable higher bandwidth without increasing DRAM chip frequencies. This paper presents a detailed perf... » read more

Pre-Silicon Verification and Validation Methodology Targeting Robust RISC-V Chip Designs (BSC)


A new technical paper, "Verification and Validation (V&V)-in-the-Loop for RISC-V Design: The Holistic Vision of BZL," was published by researchers at Barcelona Supercomputing Center. Abstract "The Barcelona Zetascale Lab (BZL) project aims to strengthening Europe's capacity in the design and manufacture of RISC-V based high-performance computing chips. In this context, we present a ho... » read more

Chip Industry Week In Review


Deals Marvell acquired Polariton Technologies, a Swiss developer of plasmonics-based silicon photonics devices. Onto Innovation is partnering with Rigaku, combining Onto’s analysis software with Rigaku’s CD-SAXS platform for advanced semiconductor process control. Onto also agreed to acquire a 27% stake in Rigaku for about $710M. Tesla plans to use Intel’s 14A process for its T... » read more

Chip Industry Technical Paper Roundup: Dec 22


New technical papers recently added to Semiconductor Engineering’s library: [table id=506 /] Find more semiconductor research papers here and in the most recent Chip Industry Week in Review. » read more

Data-Centric ML Compiler For PIM (U. of Toronto, Barcelona Supercomputing Center, ETH Zurich, Max Planck)


A new technical paper titled "A Tensor Compiler for Processing-In-Memory Architectures" was published by researchers at University of Toronto, Barcelona Supercomputing Center, ETH Zurich, and the Max Planck Institute for Software Systems. Abstract "Processing-In-Memory (PIM) devices integrated with high-performance Host processors (e.g., GPUs) can accelerate memory-intensive kernels in Ma... » read more

Chip Industry Technical Paper Roundup: Apr. 1


New technical papers recently added to Semiconductor Engineering’s library: [table id=416 /] Find more semiconductor research papers here. » read more

GPU Analysis Identifying Performance Bottlenecks That Cause Throughput Plateaus In Large-Batch Inference


A new technical paper titled "Mind the Memory Gap: Unveiling GPU Bottlenecks in Large-Batch LLM Inference" was published by researchers at Barcelona Supercomputing Center, Universitat Politecnica de Catalunya, and IBM Research. Abstract "Large language models have been widely adopted across different tasks, but their auto-regressive generation nature often leads to inefficient resource util... » read more

← Older posts