Listar GI-GAC - Congresos, conferencias, etc. por autor "Andrade, Diego"
Mostrando ítems 1-5 de 5
-
Guiding the Optimization of Parallel Codes on Multicores Using an Analytical Cache Model
Andrade, Diego; Fraguela, Basilio B.; Doallo, Ramón (2018)[Abstract]: Cache performance is particularly hard to predict in modern multicore processors as several threads can be concurrently in execution, and private cache levels are combined with shared ones. This paper presents ... -
Probing the Efficacy of Hardware-Aware Weight Pruning to Optimize the SpMM routine on Ampere GPUs
López Castro, Roberto; Andrade, Diego; Fraguela, Basilio B. (Institute of Electrical and Electronics Engineers, 2022)[Abstract]: The Deep Learning (DL) community found in pruning techniques a good way to reduce the models' resource and energy consumption. These techniques lead to smaller sparse models, but sparse computations in GPUs ... -
The New UPC++ DepSpawn High Performance Library for Data-Flow Computing with Hybrid Parallelism
Fraguela, Basilio B.; Andrade, Diego (Springer, 2022)[Abstract] Data-flow computing is a natural and convenient paradigm for expressing parallelism. This is particularly true for tools that automatically extract the data dependencies among the tasks while allowing to exploit ... -
Using Artificial Vision Techniques for Individual Player Tracking in Sport Events
López Castro, Roberto; Andrade, Diego (M D P I AG, 2019-07-31)[Abstract] We introduce a hybrid approach that can track an individual football player in a video sequence. This solution achieves a good balance between speed and accuracy, combining traditional object tracking techniques ... -
VENOM: A Vectorized N:M Format for Unleashing the Power of Sparse Tensor Cores
López Castro, Roberto; Ivanov, Andrei; Andrade, Diego; Ben-Nun, Tal; Fraguela, Basilio B.; Hoefler, Torsten (Association for Computing Machinery, 2023-11)[Abstract]: The increasing success and scaling of Deep Learning models demands higher computational efficiency and power. Sparsification can lead to both smaller models as well as higher compute efficiency, and accelerated ...