Key papers in Hardware and Architecture

Pofolia’s corpus holds 83 papers from the Hardware and Architecture subfield (2012–2024). The list below starts with the most cited.

Most cited

Ranked by citation count. Because citations accumulate over time, this list naturally leans towards work published a few years ago; for where the field is now, see “recently added”.

  • fastp: an ultra-fast all-in-one FASTQ preprocessor

    Bioinformatics · 2018 · Q1 · SJR 2.00 · FWCI 837.78 · 30,201 citations · Open access

    A new C++-based tool called fastp has been developed to perform quality control and preprocessing of FASTQ files, integrating multiple operations into a single scan.

    Go to source

  • PyTorch: An Imperative Style, High-Performance Deep Learning Library

    arXiv (Cornell University) · 2019 · 16,191 citations · Open access

    PyTorch demonstrates that deep learning frameworks can achieve both usability and high performance, offering an imperative, Pythonic style that simplifies debugging and integration while efficiently utilizing hardware accelerators like GPUs.

    Go to source

  • TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems

    arXiv (Cornell University) · 2016 · 9,793 citations · Open access

    TensorFlow is a new open-source interface and implementation for expressing and executing machine learning algorithms across diverse hardware, from mobile devices to large-scale distributed systems.

    Go to source

  • TensorFlow: A system for large-scale machine learning

    arXiv (Cornell University) · 2016 · 8,829 citations · Open access

    TensorFlow is a new machine learning system designed for large-scale, heterogeneous environments, utilizing dataflow graphs to represent computation and operations across multiple machines and devices like CPUs, GPUs, and TPUs.

    Go to source

  • Julia: A Fresh Approach to Numerical Computing

    SIAM Review · 2017 · Q1 · SJR 2.00 · FWCI 358.65 · 6,330 citations

    Julia is a new programming language designed to be both easy to use and fast, challenging the long-held belief that high-level dynamic programs must be slow.

    Go to source

  • In-Datacenter Performance Analysis of a Tensor Processing Unit

    2017 · 4,441 citations

    A custom ASIC, the Tensor Processing Unit (TPU), deployed in datacenters since 2015, accelerates neural network inference, achieving 15x-30x speedup and 30x-80x higher TOPS/Watt compared to contemporary CPUs and GPUs.

    Go to source

  • The nesC language

    ACM SIGPLAN Notices · 2014 · FWCI 85.38 · 1,653 citations

    nesC is a new programming language designed specifically for networked embedded systems, such as sensor networks, addressing their unique constraints of low power and memory.

    Go to source

  • Numba

    2015 · 1,518 citations

    Numba is a just-in-time compiler for Python that translates a subset of the language into efficient machine code for scientific and array-oriented computing.

    Go to source

  • Revolutionizing Supply Chain Management: Real-time Data Processing and Concurrency

    International Journal of Innovative Science and Research Technology (IJISRT) · 2024 · FWCI 565.18 · 1,322 citations · Open access

    This paper introduces a novel supply chain management system using Kafka for event streaming and Akka for concurrent applications to enhance data integration and real-time decision-making.

    Go to source

  • In-Datacenter Performance Analysis of a Tensor Processing Unit

    ACM SIGARCH Computer Architecture News · 2017 · FWCI 174.38 · 1,317 citations · Open access

    A custom Tensor Processing Unit (TPU) deployed in datacenters since 2015 offers a 15x-30x performance improvement and 30x-80x higher TOPS/Watt compared to contemporary CPUs and GPUs for neural network inference.

    Go to source

  • Physical Unclonable Functions and Applications: A Tutorial

    Proceedings of the IEEE · 2014 · Q1 · SJR 5.00 · FWCI 82.15 · 1,307 citations · Open access

    This tutorial introduces Physical Unclonable Functions (PUFs) as a low-cost alternative to conventional secure memories for authentication and key generation.

    Go to source

  • Introduction to Embedded Systems - A Cyber-Physical Systems Approach

    2013 · FWCI 143.51 · 1,179 citations

    This book introduces embedded systems, the ubiquitous, less visible computers controlling physical processes in everything from cars to cell phones, using a cyber-physical systems approach.

    Go to source

  • A Survey of Techniques for Approximate Computing

    ACM Computing Surveys · 2016 · Q1 · SJR 5.00 · FWCI 150.77 · 1,042 citations

    This survey categorizes and analyzes various techniques for approximate computing, which sacrifices computation quality for efficiency, highlighting its growing necessity due to increasing performance demands and limited resources.

    Go to source

  • Theano: new features and speed improvements

    arXiv (Cornell University) · 2012 · 1,040 citations · Open access

    Theano, a linear algebra compiler, has been updated with new features and significant speed enhancements.

    Go to source

  • cuDNN: Efficient Primitives for Deep Learning

    arXiv (Cornell University) · 2014 · 1,028 citations · Open access

    A new library, cuDNN, provides optimized routines for deep learning computational kernels, analogous to BLAS for HPC, to address the challenge of reoptimizing code for evolving parallel architectures.

    Go to source

  • Industrie 4.0: Hit or Hype? [Industry Forum]

    IEEE Industrial Electronics Magazine · 2014 · Q1 · SJR 1.00 · FWCI 71.54 · 981 citations

    Industrie 4.0, a concept for the fourth industrial revolution, is often understood as applying cyber-physical systems to industrial production, but its broad definitions and overambitious marketing have caused confusion.

    Go to source

  • Algorithms+Data Structures = Programs

    Cambridge University Press eBooks · 2024 · FWCI 9.50 · 962 citations

    This undergraduate textbook for CS2 courses offers a hands-on approach to mastering fundamental programming skills, data structures, and algorithms.

    Go to source

  • Chisel

    2012 · 911 citations

    Chisel is a new hardware construction language embedded in Scala, enabling advanced hardware design through parameterized generators and layered domain-specific languages.

    Go to source

  • Heterogeneous parallelization and acceleration of molecular dynamics simulations in GROMACS

    The Journal of Chemical Physics · 2020 · Q1 · FWCI 58.28 · 910 citations · Open access

    GROMACS now efficiently utilizes both GPUs and CPUs for molecular dynamics simulations through a novel heterogeneous parallelization strategy, significantly accelerating computations.

    Go to source

  • Security and privacy challenges in industrial internet of things

    2015 · 902 citations

    Industrial IoT systems, crucial for applications like critical infrastructure and modern vehicles, generate vast amounts of sensitive data, making them prime targets for cyberattacks that can cause physical damage and endanger lives.

    Go to source

Recently added

  • Algorithms+Data Structures = Programs

    Cambridge University Press eBooks · 2024 · FWCI 9.50 · 962 citations

    This undergraduate textbook for CS2 courses offers a hands-on approach to mastering fundamental programming skills, data structures, and algorithms.

    Go to source

  • Revolutionizing Supply Chain Management: Real-time Data Processing and Concurrency

    International Journal of Innovative Science and Research Technology (IJISRT) · 2024 · FWCI 565.18 · 1,322 citations · Open access

    This paper introduces a novel supply chain management system using Kafka for event streaming and Akka for concurrent applications to enhance data integration and real-time decision-making.

    Go to source

  • PyTorch 2: Faster Machine Learning Through Dynamic Python Bytecode Transformation and Graph Compilation

    2024 · 606 citations

    PyTorch 2's torch.compile feature, powered by TorchDynamo and TorchInductor, achieves 2.27x faster inference and 1.41x faster training on GPUs by dynamically compiling Python bytecode into optimized graphs.

    Go to source

  • Kokkos 3: Programming Model Extensions for the Exascale Era

    IEEE Transactions on Parallel and Distributed Systems · 2021 · Q1 · SJR 1.00 · FWCI 62.51 · 472 citations · Open access

    Kokkos version 3 introduces novel abstractions including hierarchical parallelism, task graphs, and arbitrary-sized atomic operations to enable single-source application development across diverse exascale computing architectures.

    Go to source

  • MLIR: Scaling Compiler Infrastructure for Domain Specific Computation

    2021 · 511 citations

    MLIR is a novel compiler infrastructure designed to reduce the cost of building domain-specific compilers and connect existing ones, facilitating code generation and optimization across diverse applications and hardware.

    Go to source

Add this field to your daily feed

Pick your interests and new work in your area arrives every day, summarised. Full summaries live in the app.

Open the app

Other subfields in the same field

All fields