category: Power & Performance

category: IoT, Security & Automotive

Survey of CXL Implementations and Standards (Intel, Microsoft)
By Technical Paper Link - 22 Jul, 2024 - Comments: 0

A new technical paper titled "An Introduction to the Compute Express Link (CXL) Interconnect" was published by researchers at Intel Corporation, Microsoft, and University of Washington. Abstract "The Compute Express Link (CXL) is an open industry-standard interconnect between processors and devices such as accelerators, memory buffers, smart network interfaces, persistent memory, and solid-... » read more

Improving Performance and Power Efficiency By Safely Eliminating Load Instruction Execution (ETH Zürich, Intel)
By Technical Paper Link - 22 Jul, 2024 - Comments: 0

A technical paper titled “Constable: Improving Performance and Power Efficiency by Safely Eliminating Load Instruction Execution” was published by researchers at ETH Zürich and Intel Corporation.  This paper earned the Best Paper Award in the International Symposium on Computer Architecture (ISCA). Abstract: "Load instructions often limit instruction-level parallelism (ILP) in modern pr... » read more

SiC Power Electronics Packaging: Floating Die Structure and Liquid Metal Fluidic Connection (Cambridge U. )
By Technical Paper Link - 17 Jul, 2024 - Comments: 0

A new technical paper titled "Liquid Metal Fluidic Connection and Floating Die Structure for Ultralow Thermomechanical Stress of SiC Power Electronics Packaging" was published by researchers at Cambridge University. Abstract "Coefficients of thermal expansion (CTE) of various materials in packaging structure layers vary largely, causing significant thermomechanical stress in power electroni... » read more

In Situ Backpropagation Strategy That Progressively Updates Neural Network Layers Directly in HW (TU Eindhoven)
By Technical Paper Link - 16 Jul, 2024 - Comments: 0

A new technical paper titled "Hardware implementation of backpropagation using progressive gradient descent for in situ training of multilayer neural networks" was published by researchers at Eindhoven University of Technology. Abstract "Neural network training can be slow and energy-expensive due to the frequent transfer of weight data between digital memory and processing units. Neuromorp... » read more

On-Chip Communication For Programmable Accelerators In Heterogeneous SoCs (Columbia, IBM)
By Technical Paper Link - 10 Jul, 2024 - Comments: 0

A technical paper titled “Towards Generalized On-Chip Communication for Programmable Accelerators in Heterogeneous Architectures” was published by researchers at Columbia University and IBM Thomas J. Watson Research Center. Abstract: "We present several enhancements to the open-source ESP platform to support flexible and efficient on-chip communication for programmable accelerators in het... » read more

Analysis Of The On-DRAM-Die Read Disturbance Mitigation Method: Per Row Activation Counting
By Technical Paper Link - 10 Jul, 2024 - Comments: 0

A technical paper titled “Understanding the Security Benefits and Overheads of Emerging Industry Solutions to DRAM Read Disturbance” was published by researchers at ETH Zürich and TOBB University of Economics and Technology. Abstract: "We present the first rigorous security, performance, energy, and cost analyses of the state-of-the-art on-DRAM-die read disturbance mitigation method, Per... » read more

2D UltraLow Temperatures, High Performance Quantum
By Technical Paper Link - 07 Jul, 2024 - Comments: 0

A new technical paper titled "Electrically tunable giant Nernst effect in two-dimensional van der Waals heterostructures" was published by researchers at EPFL and National Institute for Materials Science (Japan). Abstract "The Nernst effect, a transverse thermoelectric phenomenon, has attracted significant attention for its potential in energy conversion, thermoelectrics and spintronics. ... » read more

Lower Energy, High Performance LLM on FPGA Without Matrix Multiplication
By Technical Paper Link - 27 Jun, 2024 - Comments: 0

A new technical paper titled "Scalable MatMul-free Language Modeling" was published by UC Santa Cruz, Soochow University, UC Davis, and LuxiTech. Abstract "Matrix multiplication (MatMul) typically dominates the overall computational cost of large language models (LLMs). This cost only grows as LLMs scale to larger embedding dimensions and context lengths. In this work, we show that MatMul... » read more

A Memory Device With MoS2 Channel For High-Density 3D NAND Flash-Based In-Memory Computing
By Technical Paper Link - 26 Jun, 2024 - Comments: 0

A technical paper titled “Low-Power Charge Trap Flash Memory with MoS2 Channel for High-Density In-Memory Computing" was published by researchers at Kyungpook National University, Sungkyunkwan University, Dankook University, and Kwangwoon University. Abstract: "With the rise of on-device artificial intelligence (AI) technology, the demand for in-memory computing has surged for data-intensiv... » read more

Thermoelectric Active Cooling Hot Spots in Chips
By Technical Paper Link - 26 Jun, 2024 - Comments: 0

A technical paper titled “Thermoelectric active cooling for transient hot spots in microprocessors” was published by researchers at the University of Pittsburgh and Carnegie Mellon University. Abstract: "Modern microprocessor performance is limited by local hot spots induced at high frequency by busy integrated circuit elements such as the clock generator. Locally embedded thermoelectric ... » read more

← Older posts

  • Sponsors

About

Navigation

Connect With Us

Copyright ©2013-2024 SMG   |  Terms of Service  |  Privacy Policy