%0 Conference Proceedings
%4 sid.inpe.br/mtc-m16c/2018/12.27.18.19
%2 sid.inpe.br/mtc-m16c/2018/12.27.18.19.43
%@issn 2179-4847
%T An efficient flash-aware spatial index for points
%D 2018
%A Carniel, Anderson Chaves,
%A Roumelis, George,
%A Ciferri, Ricardo Rodrigues,
%A Vassilakopoulos, Michael,
%A Corral, Antonio,
%A Ciferri, Cristina Dutra de Aguiar,
%@affiliation Universidade de São Paulo (USP)
%@affiliation University of Thessaly
%@affiliation Universidade Federal de São Carlos (UFSCar)
%@affiliation University of Thessaly
%@affiliation University of Almeria
%@affiliation Universidade de São Paulo (USP)
%@electronicmailaddress accarniel@gmail.com
%@electronicmailaddress groumelis@uth.gr
%@electronicmailaddress ricardo@dc.ufscar.br
%@electronicmailaddress mvasilako@uth.gr
%@electronicmailaddress acorral@ual.es
%@electronicmailaddress cdac@icmc.usp.br
%E Vinhas, Lúbia (INPE),
%E Campelo, Claudio (UFCG),
%B Simpósio Brasileiro de Geoinformática, 19 (GEOINFO)
%C Campina Grande
%8 05-07 dez. 2018
%I Instituto Nacional de Pesquisas Espaciais (INPE)
%J São José dos Campos
%P 68-79
%X Spatial database systems often employ spatial indices to speed up the processing of spatial queries. In addition, modern spatial database applica- tions are interested in exploiting the positive characteristics of flash-based Solid State Drives (SSDs) like fast reads and writes. However, designing spatial in- dices for SSDs (i.e., flash-aware spatial indices) has been a challenging task because of the intrinsic characteristics of these devices. In this paper, we pro- pose the eFIND xBR+-tree, a novel flash-aware spatial index for points. The eFIND xBR+-tree combines the efficient indexing method of the xBR+-tree with the sophisticated data structures and algorithms of eFIND to handle points in SSDs efficiently. Experiments carried out considering real and synthetic spa- tial data showed that the eFIND xBR+-tree overcame its closest competitor by reducing the elapsed time to construct the index from 28.4% to 83.5% and to execute spatial queries up to 34.6%.
%@language pt
%3 p7.pdf