Suchen

Filter löschen
Filtern nach: Committee Member Nguyen, Thinh Filter Committee Member: Nguyen, Thinh entfernen Committee Member Bose, Bella Filter Committee Member: Bose, Bella entfernen Degree Level Doctoral Filter Degree Level: Doctoral entfernen

Suchergebnisse

  • Architecture Optimizations for Memory Systems of Throughput Processors

    Schöpfer:
    Gu, Yongbin
    Abstract:
    Throughput-oriented processors, such as graphics processing units (GPUs), have been increasingly used to accelerate general purpose computing, including machine learning models that are being utilized in numerous disciplines. Thousands of concurrently running threads in a GPU demand a highly efficient memory subsystem for data supply in GPUs. In this dissertation,...
    Resource Type:
    Dissertation

  • Adaptive Streaming Methods using Dual Protocols and Reinforcement Learning

    Schöpfer:
    Gatimu, Kevin
    Abstract:
    The ubiquity of high quality video and proliferation of mobile devices has contributed to an unprecedented rise in video consumption. HTTP, in conjunction with adaptive streaming, has become the de facto mechanism for delivering the vast majority of video as it readily caters to heterogeneous networks and devices. This dissertation...
    Resource Type:
    Dissertation

  • Dynamic and Dual Streaming Methods for H.264 Video and Parallel Performance Modeling

    Schöpfer:
    Johnson, Taylor N.
    Abstract:
    Traditional approaches to streaming H.264 video over a network typically rely on a single method of transport (i.e., reliable or unreliable) and/or use static values for parameters that can have a significant negative impact on the perceptual quality of the received video. This dissertation presents a dynamic method for wireless...
    Resource Type:
    Dissertation

  • Offloading of I/O transactions in current CPU architectures

    Schöpfer:
    Larsen, Steen K.
    Abstract:
    IO transactions within a computer system have evolved along with other system components (i.e., CPU, memory, video) from programmed IO (PIO). In current mainstream systems (spanning from HPC to mobile) the IO transactions are CPU-centric descriptor-based DMA transactions. The key benefit is that slower IO devices can DMA write system...
    Resource Type:
    Dissertation

  • Security protocols for linear network coding

    Schöpfer:
    Adeli, Majid
    Abstract:
    Network coding, as the next generation of data routing protocols, enables each intermediate node in a network to process and encode its received data before forwarding it to the next nodes. Hence, the core idea in network coding is to allow a network to encode the data that is being...
    Resource Type:
    Dissertation

  • The dynamic speculation and performance prediction of parallel loops

    Schöpfer:
    Zier, David A.
    Abstract:
    General purpose computer systems have seen increased performance potential through the parallel processing capabilities of multicore processors. Yet this potential performance can only be attained through parallel applications, thus forcing software developers to rethink how everyday applications are designed. The most readily form of Thread Level Parallelism (TLP) within any...
    Resource Type:
    Dissertation

  • Low-complexity high-speed VLSI design of low-density parity-check decoders

    Schöpfer:
    Cui, Zhiqiang
    Abstract:
    Low-Density Parity-check (LDPC) codes have attracted considerable attention due to their capacity approaching performance over AWGN channel and highly parallelizable decoding schemes. They have been considered in a variety of industry standards for the next generation communication systems. In general, LDPC codes achieve outstanding performance with large codeword lengths (e.g.,...
    Resource Type:
    Dissertation