Towards a learning-based performance modeling for accelerating Deep Neural Networks

Autor: Damiano Perri, Osvaldo Gervasi, Sergio Tasso, Paolo Sylos Labini, Flavio Vella

Publikováno v: Computational Science and Its Applications – ICCSA 2019 ISBN: 9783030242886
ICCSA (1)

Emerging applications such as Deep Learning are often data-driven, thus traditional approaches based on auto-tuners are not performance effective across the wide range of inputs used in practice. In the present paper, we start an investigation of pre

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::768c184ff0ed40f7e1bf7b6d8f02a46a
http://arxiv.org/abs/2212.05031

ProbGraph: High-Performance and High-Accuracy Graph Mining with Probabilistic Set Representations

Autor: Maciej Besta, Cesare Miglioli, Paolo Sylos Labini, Jakub Tětek, Patrick Iff, Raghavendra Kanakagiri, Saleh Ashkboos, Kacper Janda, Michał Podstawski, Grzegorz Kwaśniewski, Niels Gleinig, Flavio Vella, Onur Mutlu, Torsten Hoefler

Publikováno v: Besta, M, Miglioli, C, Labini, P S, Tetek, J, Iff, P, Kanakagiri, R, Ashkboos, S, Janda, K, Podstawski, M, Kwasniewski, G, Gleinig, N, Vella, F, Mutlu, O & Hoefler, T 2022, ProbGraph : High-Performance and High-Accuracy Graph Mining with Probabilistic Set Representations . in Proceedings of SC 2022 : International Conference for High Performance Computing, Networking, Storage and Analysis . IEEE, International Conference for High Performance Computing, Networking, Storage and Analysis, SC, vol. 2022-November, pp. 1-17, 2022 International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2022, Dallas, United States, 13/11/2022 . https://doi.org/10.1109/SC41404.2022.00048

Important graph mining problems such as Clustering are computationally demanding. To significantly accelerate these problems, we propose ProbGraph: a graph representation that enables simple and fast approximate parallel graph mining with strong theo

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::53505069ab1f212588a4bd5c730f6566

Blocking Sparse Matrices to Leverage Dense-Specific Multiplication

Autor: Paolo Sylos Labini, Massimo Bernaschi, Werner Nutt, Francesco Silvestri, Flavio Vella

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::4daf667128cd5f23a2d7ce4af8a72588
https://hdl.handle.net/11577/3470280

Algorithm Design for Tensor Units

Autor: Rezaul Chowdhury, Francesco Silvestri, Flavio Vella

Publikováno v: Euro-Par 2021: Parallel Processing ISBN: 9783030856649
Euro-Par

To respond to the intense computational load of deep neural networks, a plethora of domain-specific architectures have been introduced, such as Google Tensor Processing Units and NVIDIA Tensor Cores. A common feature of these architectures is a hardw

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e9e892a1a3ed91436fae42e3e7fd3671
http://hdl.handle.net/11577/3413090

Multilevel Parallelism for the Exploration of Large-Scale Graphs

Autor: Enrico Mastrostefano, Flavio Vella, Massimo Bernaschi, Mauro Bisson

Publikováno v: IEEE Transactions on Multi-Scale Computing Systems 4 (2018): 204–215. doi:10.1109/TMSCS.2018.2797195
info:cnr-pdr/source/autori:Bernaschi M.; Bisson M.; Mastrostefano E.; Vella F./titolo:Multilevel parallelism for the exploration of large-scale graphs/doi:10.1109%2FTMSCS.2018.2797195/rivista:IEEE Transactions on Multi-Scale Computing Systems/anno:2018/pagina_da:204/pagina_a:215/intervallo_pagine:204–215/volume:4

We present the most recent release of our parallel implementation of the BFS and BC algorithms for the study of large scale graphs. Although our reference platform is a high-end cluster of new generation NVIDIA GPUs and some of our optimizations are

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b59c245c98ddcd65f28b9025d5224b4c
https://doi.org/10.1109/tmscs.2018.2797195