• No se han encontrado resultados

Comprehensive evaluation of a New GPU based approach to the shortest path problem

N/A
N/A
Protected

Academic year: 2020

Share "Comprehensive evaluation of a New GPU based approach to the shortest path problem"

Copied!
20
0
0

Texto completo

Loading

Figure

Fig. 1 Dijkstra’s algorithm steps: Initialization (a), edge relaxation (b), and settlement (c).
Table 1 Values selected for threadblock-size and L1 cache state through kernel characteri- characteri-zation process for Fermi (F) and Kepler (K)
Fig. 2 Experimental results of Mart´ın’s graphs: Execution times of CPU-Mart´ın vs. our CPU-Crauser (a), GPU-Mart´ın vs
Fig. 3 Scenario A: Execution times of successors algorithmic variants (GPU Mart´ın, CPU Crauser, and GPU Crauser) for synthetic random graphs.
+7

Referencias

Documento similar

Government policy varies between nations and this guidance sets out the need for balanced decision-making about ways of working, and the ongoing safety considerations

No obstante, como esta enfermedad afecta a cada persona de manera diferente, no todas las opciones de cuidado y tratamiento pueden ser apropiadas para cada individuo.. La forma

Linked data, enterprise data, data models, big data streams, neural networks, data infrastructures, deep learning, data mining, web of data, signal processing, smart cities,

Para el estudio de este tipo de singularidades en el problema de Muskat, consideraremos el caso unif´ asico, esto es, un ´ unico fluido en el vac´ıo con µ 1 = ρ 1 = 0.. Grosso modo,

Due to the the low amount of avail- able training data in a signature real case, Sequential For- ward Feature Selection (SFFS) algorithm [13] is performed in order to obtain a subset

In the previous sections we have shown how astronomical alignments and solar hierophanies – with a common interest in the solstices − were substantiated in the

Correlation between structural complexity, proportion of species included in the Red Lists at world and national level, proportion of threatened species included

CHAPTER 5. EVALUATION AND RESULTS 35.. In fact, in all variants of the algorithm, the sequential executions outperform the single work-item executions. The reason for this can be