Zobrazeno 21 - 30
of 115
pro vyhledávání: '"Besta, Maciej"'
Autor:
Gianinazzi, Lukas, Ben-Nun, Tal, Besta, Maciej, Ashkboos, Saleh, Baumann, Yves, Luczynski, Piotr, Hoefler, Torsten
We present a new parallel model of computation suitable for spatial architectures, for which the energy used for communication heavily depends on the distance of the communicating processors. In our model, processors have locations on a conceptual tw
Externí odkaz:
http://arxiv.org/abs/2205.04934
Triangle count and local clustering coefficient are two core metrics for graph analysis. They find broad application in analyses such as community detection and link recommendation. Current state-of-the-art solutions suffer from synchronization overh
Externí odkaz:
http://arxiv.org/abs/2202.13976
Publikováno v:
SPAA '21: 33rd ACM Symposium on Parallelism in Algorithms and Architectures, Virtual Event, USA, 6-8 July, 2021, 243-253
We present a parallel k-clique listing algorithm with improved work bounds (for the same depth) in sparse graphs with low degeneracy or arboricity. We achieve this by introducing and analyzing a new pruning criterion for a backtracking search. Our al
Externí odkaz:
http://arxiv.org/abs/2109.09663
Autor:
Kwasniewski, Grzegorz, Kabić, Marko, Ben-Nun, Tal, Ziogas, Alexandros Nikolaos, Saethre, Jens Eirik, Gaillard, André, Schneider, Timo, Besta, Maciej, Kozhevnikov, Anton, VandeVondele, Joost, Hoefler, Torsten
Publikováno v:
Published at Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, November, 2021(SC'21)
Matrix factorizations are among the most important building blocks of scientific computing. State-of-the-art libraries, however, are not communication-optimal, underutilizing current parallel architectures. We present novel algorithms for Cholesky an
Externí odkaz:
http://arxiv.org/abs/2108.09337
Autor:
Gianinazzi, Lukas, Fries, Maximilian, Dryden, Nikoli, Ben-Nun, Tal, Besta, Maciej, Hoefler, Torsten
We present a novel neural architecture to solve graph optimization problems where the solution consists of arbitrary node labels, allowing us to solve hard problems like graph coloring. We train our model using reinforcement learning, specifically po
Externí odkaz:
http://arxiv.org/abs/2106.03594
The growing size of data center and HPC networks pose unprecedented requirements on the scalability of simulation infrastructure. The ability to simulate such large-scale interconnects on a simple PC would facilitate research efforts. Unfortunately,
Externí odkaz:
http://arxiv.org/abs/2105.12663
Autor:
Besta, Maciej, Grob, Raphael, Miglioli, Cesare, Bernold, Nicola, Kwasniewski, Grzegorz, Gjini, Gabriel, Kanakagiri, Raghavendra, Ashkboos, Saleh, Gianinazzi, Lukas, Dryden, Nikoli, Hoefler, Torsten
Publikováno v:
Proceedings of the 28th SIGKDD Conference on Knowledge Discovery and Data Mining (KDD'22), 2022
Link prediction is one of the central problems in graph mining. However, recent studies highlight the importance of higher-order network analysis, where complex structures called motifs are the first-class citizens. We first show that existing link p
Externí odkaz:
http://arxiv.org/abs/2106.00761
Autor:
Kwasniewski, Grzegorz, Ben-Nun, Tal, Gianinazzi, Lukas, Calotoiu, Alexandru, Schneider, Timo, Ziogas, Alexandros Nikolaos, Besta, Maciej, Hoefler, Torsten
Determining I/O lower bounds is a crucial step in obtaining communication-efficient parallel algorithms, both across the memory hierarchy and between processors. Current approaches either study specific algorithms individually, disallow programmatic
Externí odkaz:
http://arxiv.org/abs/2105.07203
Autor:
Besta, Maciej, Kanakagiri, Raghavendra, Kwasniewski, Grzegorz, Ausavarungnirun, Rachata, Beránek, Jakub, Kanellopoulos, Konstantinos, Janda, Kacper, Vonarburg-Shmaria, Zur, Gianinazzi, Lukas, Stefan, Ioana, Luna, Juan Gómez, Copik, Marcin, Kapp-Schwoerer, Lukas, Di Girolamo, Salvatore, Konieczny, Marek, Blach, Nils, Mutlu, Onur, Hoefler, Torsten
Simple graph algorithms such as PageRank have been the target of numerous hardware accelerators. Yet, there also exist much more complex graph mining algorithms for problems such as clustering or maximal clique listing. These algorithms are memory-bo
Externí odkaz:
http://arxiv.org/abs/2104.07582
Autor:
Besta, Maciej, Vonarburg-Shmaria, Zur, Schaffner, Yannick, Schwarz, Leonardo, Kwasniewski, Grzegorz, Gianinazzi, Lukas, Beranek, Jakub, Janda, Kacper, Holenstein, Tobias, Leisinger, Sebastian, Tatkowski, Peter, Ozdemir, Esref, Balla, Adrian, Copik, Marcin, Lindenberger, Philipp, Kalvoda, Pavel, Konieczny, Marek, Mutlu, Onur, Hoefler, Torsten
Publikováno v:
International Conference on Very Large Data Bases (VLDB), 2021
We propose GraphMineSuite (GMS): the first benchmarking suite for graph mining that facilitates evaluating and constructing high-performance graph mining algorithms. First, GMS comes with a benchmark specification based on extensive literature review
Externí odkaz:
http://arxiv.org/abs/2103.03653