Point Cloud Processing on Apache Spark platform (CROSBI ID 638973)
Prilog sa skupa u zborniku | izvorni znanstveni rad | međunarodna recenzija
Podaci o odgovornosti
Pajić, Vladimir ; Govedarica, Miro ; Galić, Zdravko ; Alargić, Ivan
hrvatski
Point Cloud Processing on Apache Spark platform
Modern geoinformation technologies for data acquisition and processing, such as laser scanning and photogrammetry, can generate point clouds which contains hunderds of millions or billions 3D points. Efficient management of such amount of data requires techniques based on cluster of computers. Apache Spark is the new software framework for parallel processing of large amounts of data. It is designed to retain scalability and fault tollerance of MapReduce and support applications which could not be implemented efficiently with MapReduce. This paper presents the results of our research on the processing of point clouds on Apache Spark framework. The main topic covered is the partition and distribution of data on a cluster of computers in order to achieve maximum efficiency. Also, the various approaches for spatial indexing of point clouds are presented because it is closely related to the problem of data partition.
Distributed processing; Cloud computing; MapReduce; Point Cloud
nije evidentirano
engleski
Point Cloud Processing on Apache Spark platform
nije evidentirano
Distributed processing; Cloud computing; MapReduce; Point Cloud
nije evidentirano
Podaci o prilogu
127-130.
2015.
nije evidentirano
objavljeno
978-86-85525-15-5
Podaci o matičnoj publikaciji
Zbornik radova YUINFO 2015
Ivković, Miodrag
Beograd: Društvo za informacione sisteme i računarske mreže
Podaci o skupu
21 Conference YUINFO 2015
predavanje
08.03.2015-11.03.2015
Kopaonik, Srbija