F
Federico Silla
Researcher at Polytechnic University of Valencia
Publications - 130
Citations - 2209
Federico Silla is an academic researcher from Polytechnic University of Valencia. The author has contributed to research in topics: Virtualization & Network on a chip. The author has an hindex of 26, co-authored 125 publications receiving 2070 citations. Previous affiliations of Federico Silla include University of Valencia & Heidelberg University.
Papers
More filters
Proceedings ArticleDOI
rCUDA: Reducing the number of GPU-based accelerators in high performance clusters
TL;DR: This paper details a framework that enables remote GPU acceleration in HPC clusters, thus allowing a reduction in the number of accelerators installed in the cluster, which leads to energy, acquisition, maintenance, and space savings.
Journal ArticleDOI
A complete and efficient CUDA-sharing solution for HPC clusters
TL;DR: The key features, architectural design, and implementation of rCUDA, an advanced framework to enable remote and transparent GPGPU acceleration in HPC clusters, are detailed and an extensive experimental evaluation of remote GPU acceleration is provided.
Proceedings ArticleDOI
Addressing Manufacturing Challenges with Cost-Efficient Fault Tolerant Routing
Samuel Rodrigo,Jose Flich,Antoni Roca,Simone Medardoni,Davide Bertozzi,J. Camacho,Federico Silla,José Duato +7 more
TL;DR: Universal Logic-Based Distributed Routing (uLBDR) as mentioned in this paper is an efficient logic-based mechanism that adapts to any irregular topology derived from 2D meshes, being an alternative to the use of routing tables.
Proceedings ArticleDOI
Enabling CUDA acceleration within virtual machines using rCUDA
José Duato,Antonio J. Peña,Federico Silla,Juan C. Fernández,Rafael Mayo,Enrique S. Quintana-Ortí +5 more
TL;DR: This paper describes the use of rCUDA, a GPGPU (General-Purpose Computation on GPUs) virtualization framework, to permit the execution of GPU-accelerated applications within virtual machines (VMs), thus enabling GPG PU capabilities on any virtualized environment.
Journal ArticleDOI
High-performance routing in networks of workstations with irregular topology
Federico Silla,José Duato +1 more
TL;DR: Evaluation results for several different tapologies and message distributions show that the new routing algorithms are able to increase throughput for random traffic by a factor of up to 4 with respect to the original up*/down* algorithm, also reducing latency significantly.