JavaScript is disabled for your browser. Some features of this site may not work without it.

    Listar

    Todo RIUMAComunidades & ColeccionesPor fecha de publicaciónAutoresTítulosMateriasTipo de publicaciónCentrosEsta colecciónPor fecha de publicaciónAutoresTítulosMateriasTipo de publicaciónCentros

    Mi cuenta

    AccederRegistro

    Estadísticas

    Ver Estadísticas de uso

    DE INTERÉS

    Datos de investigaciónReglamento de ciencia abierta de la UMAPolítica de RIUMAPolitica de datos de investigación en RIUMASHERPA/RoMEODulcinea
    Preguntas frecuentesManual de usoDerechos de autorContacto/Sugerencias
    Ver ítem 
    •   RIUMA Principal
    • Investigación
    • Arquitectura de Computadores - (AC)
    • AC - Contribuciones a congresos científicos
    • Ver ítem
    •   RIUMA Principal
    • Investigación
    • Arquitectura de Computadores - (AC)
    • AC - Contribuciones a congresos científicos
    • Ver ítem

    Improvements in Hardware Transactional Memory for GPU Architectures

    • Autor
      Villegas, Alejandro; Asenjo-Plaza, RafaelAutoridad Universidad de Málaga; Navarro, Ángeles; Plata-Gonzalez, Oscar GuillermoAutoridad Universidad de Málaga
    • Fecha
      2016-07-20
    • Palabras clave
      Ordenadores - Equipo de entrada y salida
    • Resumen
      In the multi-core CPU world, transactional memory (TM)has emerged as an alternative to lock-based programming for thread synchronization. Recent research proposes the use of TM in GPU architectures, where a high number of computing threads, organized in SIMT fashion, requires an effective synchronization method. In contrast to CPUs, GPUs offer two memory spaces: global memory and local memory. The local memory space serves as a shared scratch-pad for a subset of the computing threads, and it is used by programmers to speed-up their applications thanks to its low latency. Prior work from the authors proposed a lightweight hardware TM (HTM) support based in the local memory, modifying the SIMT execution model and adding a conflict detection mechanism. An efficient implementation of these features is key in order to provide an effective synchronization mechanism at the local memory level. After a quick description of the main features of our HTM design for GPU local memory, in this work we gather together a number of proposals designed with the aim of improving those mechanisms with high impact on performance. Firstly, the SIMT execution model is modified to increase the parallelism of the application when transactions must be serialized in order to make forward progress. Secondly, the conflict detection mechanism is optimized depending on application characteristics, such us the read/write sets, the probability of conflict between transactions and the existence of read-only transactions. As these features can be present in hardware simultaneously, it is a task of the compiler and runtime to determine which ones are more important for a given application. This work includes a discussion on the analysis to be done in order to choose the best configuration solution.
    • URI
      http://hdl.handle.net/10630/11858
    • Compartir
      RefworksMendeley
    Mostrar el registro completo del ítem
    Ficheros
    paper-19.pdf (297.9Kb)
    Colecciones
    • AC - Contribuciones a congresos científicos

    Estadísticas

    Ver Estadísticas de uso
    Buscar en Dimension
    REPOSITORIO INSTITUCIONAL UNIVERSIDAD DE MÁLAGA
    REPOSITORIO INSTITUCIONAL UNIVERSIDAD DE MÁLAGA
     

     

    REPOSITORIO INSTITUCIONAL UNIVERSIDAD DE MÁLAGA
    REPOSITORIO INSTITUCIONAL UNIVERSIDAD DE MÁLAGA