ÇѱÛÁ¦¸ñ(Korean Title) |
GPU¸¦ È°¿ëÇÑ R*-tree¿¡¼ÀÇ ºÎºÐ ³ëµå º´·Ä ó¸® ¹æ¹ý |
¿µ¹®Á¦¸ñ(English Title) |
A Parallel Processing Method for Partial Nodes in R*-tree Using GPU |
ÀúÀÚ(Author) |
񊬧
¿Àº´¿ì
Seong Kim
Byoung-Woo Oh
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 20 NO. 06 PP. 0139 ~ 0144 (2012. 12) |
Çѱ۳»¿ë (Korean Abstract) |
°ø°£ µ¥ÀÌÅÍ Ã³¸®´Â GIS, ÅÚ·¹¸Åƽ½º µî ±¤¹üÀ§ÇÑ ºÐ¾ß¿¡¼ ³Î¸® »ç¿ëµÇ°í ÀÖ´Ù. ±×·¯³ª ÇöÀç »ç¿ëµÇ°í ÀÖ´Â °ø°£ µ¥ÀÌÅÍ ÁúÀÇ Ã³¸® ±â¹ýÀº CPU¸¦ »ç¿ëÇÏ¿© ¼øÂ÷ÀûÀ¸·Î ÁúÀÇ Ã³¸®¸¦ ¼öÇàÇϹǷΠÁúÀÇ Ã³¸® ½Ã°£ÀÌ »ó´ëÀûÀ¸·Î ¿À·¡ °É¸°´Ù´Â ´ÜÁ¡ÀÌ Á¸ÀçÇÑ´Ù. ±×·¯³ª °ø°£ µ¥ÀÌÅÍ ÁúÀÇ Ã³¸®¸¦ º´·Ä·Î ¼öÇàÇßÀ» ¶§ ó¸® ½Ã°£À» ÁÙÀÌ´Â °ÍÀÌ °¡´ÉÇÏ´Ù. µû¶ó¼ º» ¿¬±¸¿¡¼´Â GPU¸¦ È°¿ëÇÏ¿© °ø°£ µ¥ÀÌÅÍ ÁúÀÇ Ã³¸®¸¦ º´·Ä·Î ¼öÇàÇÏ´Â ¿¬±¸¸¦ ÁøÇàÇÑ´Ù. ¶ÇÇÑ, CPU¸¦ ÀÌ¿ëÇÏ¿© ÁúÀÇ Ã³¸®¸¦ ¼öÇàÇÑ °á°ú¿Í ºñ±³ÇÏ¿© ¼Óµµ Çâ»ó Á¤µµ¿¡ ´ëÇÑ °á°ú¸¦ Á¦½ÃÇÑ´Ù.
|
¿µ¹®³»¿ë (English Abstract) |
The R*-tree manages hierarchical nodes for efficient access of spatial data. We propose a method that maintains partial nodes of R*-tree in the GPU memory to im prove efficiency using parallel processing. The proposed method attempts to load as many nodes as possible to the GPU memory. The new nodes are inserted to manage the rest of R*-tree nodes in the main memory. The experimental result shows that the proposed method is more efficient than the main memory based R*-tree.
|
Å°¿öµå(Keyword) |
R*-tree
GPU
À妽º
°ø°£ µ¥ÀÌÅÍ
º´·Ä ó¸®
R*-tree
GPU
Index
Spatial Data
Parallel Processing
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|