Now showing items 1-20 of 30

  • Adapting deep neural networks to a low-power environment 

    Mañas Sánchez, Oscar (Universitat Politècnica de Catalunya, 2017)
    Bachelor thesis
    Open Access
    These days, working with deep neural networks goes hand in hand with the use of GPUs. Once a deep neural network has been trained for hours, days, or even weeks on a desktop GPU, it is deployed in the field where it runs ...
  • An enegy-efficient FPGA accelerator for convolutional neural networks 

    Delgado Ventosa, Berta (Universitat Politècnica de Catalunya, 2018-10-25)
    Master thesis
    Open Access
    This project focuses on a state-of-the-art DNN specifically build for image clas sification. We develop a new architecture design that will run on an experimental Intel accelerator platform called HARP. We obtain low-power ...
  • Architectural support for high-performing hardware transactional memory systems 

    Lupon Navazo, Marc (Universitat Politècnica de Catalunya, 2011-12-23)
    Doctoral thesis
    Open Access
    Parallel programming presents an efficient solution to exploit future multicore processors. Unfortunately, traditional programming models depend on programmer’s skills for synchronizing concurrent threads, which makes ...
  • Architecture support for intrusion detection systems 

    Sreekar Shenoy, Govind (Universitat Politècnica de Catalunya, 2012-10-30)
    Doctoral thesis
    Open Access
    System security is a prerequisite for efficient day-to-day transactions. As a consequence, Intrusion Detection Systems (IDS) are commonly used to provide an effective security ring to systems in a network. An IDS operates ...
  • Characterization of Speech Recognition Systems on GPU Architectures 

    Segura Salvador, Albert (Universitat Politècnica de Catalunya, 2016-07-04)
    Master thesis
    Open Access
    This master thesis characterizes the performance and energy bottlenecks of speech recognition systems when running on modern GPU, with the aim of providing useful information for designing future GPU architectures, as well ...
  • Code optimizations for narrow bitwidth architectures 

    Bhagat, Indu (Universitat Politècnica de Catalunya, 2012-02-23)
    Doctoral thesis
    Open Access
    This thesis takes a HW/SW collaborative approach to tackle the problem of computational inefficiency in a holistic manner. The hardware is redesigned by restraining the datapath to merely 16-bit datawidth (integer datapath ...
  • Co-designed solutions for overhead removal in dynamically typed languages 

    Dot Artigas, Gem (Universitat Politècnica de Catalunya, 2016-07-26)
    Doctoral thesis
    Open Access
    Dynamically typed languages are ubiquitous in today's applications. These languages ease the task of programmers but introduce significant runtime overheads since variables are neither declared nor bound to a particular ...
  • Comparative Study of Prefetching Mechanisms 

    Torrents Lapuerta, Martí (Universitat Politècnica de Catalunya, 2009-09-23)
    Master thesis
    Restricted access - confidentiality agreement
  • Design of a distributed memory unit for clustered microarchitectures 

    Bieschewski, Stefan (Universitat Politècnica de Catalunya, 2013-06-20)
    Doctoral thesis
    Open Access
    Power constraints led to the end of exponential growth in single–processor performance, which characterized the semiconductor industry for many years. Single–chip multiprocessors allowed the performance growth to continue ...
  • Design of Clustered Superscalar Microarchitectures 

    Parcerisa Bundó, Joan Manuel (Universitat Politècnica de Catalunya, 2004-06-17)
    Doctoral thesis
    Open Access
    L'objectiu d'aquesta tesi és proposar noves tècniques per al disseny de microarquitectures clúster superescalars eficients. Les microarquitectures clúster particionen el disseny de diversos components crítics del hardware ...
  • Energy Efficient Accelerator for Self Driving Cars 

    Syeda, Azmath (Universitat Politècnica de Catalunya, 2018-10-15)
    Master thesis
    Restricted access - confidentiality agreement
  • Fòrum d'Informàtica Universitat Empresa : 4 i 5 de maig : microprocessadores actuales 

    Facultat d'Informàtica de Barcelona; Junior Empresa d'Informàtica; Universitat Politècnica de Catalunya. Institut de Ciències de l'Educació (1994-05)
    Audiovisual
    Open Access
  • HW/SW mechanisms for instruction fusion, issue and commit in modern u-processors 

    Deb, Abhishek (Universitat Politècnica de Catalunya, 2012-05-03)
    Doctoral thesis
    Open Access
    In this thesis we have explored the co-designed paradigm to show alternative processor design points. Specifically, we have provided HW/SW mechanisms for instruction fusion, issue and commit for modern processors. We have ...
  • Low-power architectures for automatic speech recognition 

    Tabani, Hamid (Universitat Politècnica de Catalunya, 2018-02-21)
    Doctoral thesis
    Open Access
    Automatic Speech Recognition (ASR) is one of the most important applications in the area of cognitive computing. Fast and accurate ASR is emerging as a key application for mobile and wearable devices. These devices, such ...
  • Managing dynamic non-uiform cache architectures 

    Lira Rueda, Javier (Universitat Politècnica de Catalunya, 2011-11-25)
    Doctoral thesis
    Open Access
    Researchers from both academia and industry agree that future CMPs will accommodate large shared on-chip last-level caches. However, the exponential increase in multicore processor cache sizes accompanied by growing on-chip ...
  • Microarchitectural simulator for shader cores in a modern GPU simulation infrastructure 

    Joseph, Diya (Universitat Politècnica de Catalunya, 2019-07-05)
    Master thesis
    Open Access
    The objective of this project is to redesign the shader cores in TEAPOT, a cycle-accurate simulator for mobile-GPU systems, validate it's functionality and accuracy and then explore the new microarchitecture by performing ...
  • Microarchitectural Techniques to Exploit Repetitive Computations and Values 

    Molina Clemente, Carlos (Universitat Politècnica de Catalunya, 2005-07-10)
    Doctoral thesis
    Open Access
    La dependencia de datos es una de las principales razones que limitan el rendimiento de los procesadores actuales. Algunos estudios han demostrado, que las aplicaciones no pueden alcanzar más de una decena de instrucciones ...
  • Mitosis based speculative multithreaded architectures 

    Madriles Gimeno, Carles (Universitat Politècnica de Catalunya, 2012-07-23)
    Doctoral thesis
    Open Access
    In the last decade, industry made a right-hand turn and shifted towards multi-core processor designs, also known as Chip-Multi-Processors (CMPs), in order to provide further performance improvements under a reasonable power ...
  • Multipath: un sistema para la programación lógica 

    Tubella Murgadas, Jordi (Universitat Politècnica de Catalunya, 1996-11-13)
    Doctoral thesis
    Open Access
  • Optimizing SIMD execution in HW/SW co-designed processors 

    Kumar, Rakesh (Universitat Politècnica de Catalunya, 2014-07-24)
    Doctoral thesis
    Open Access
    SIMD accelerators are ubiquitous in microprocessors from different computing domains. Their high compute power and hardware simplicity improve overall performance in an energy efficient manner. Moreover, their replicated ...