Çѱ¹Á¤º¸Åë½ÅÇÐȸ ³í¹®Áö (Journal of the Korea Institute of Information and Communication Engineering)
ÇѱÛÁ¦¸ñ(Korean Title) |
º´·Ä ÇÁ·Î±×·¥¿¡¼ÀÇ È¿À²ÀûÀÎ ´ë¿ë·® ÆÄÀÏ ÀÔÃâ·Â ¹æ½ÄÀÇ ºñ±³ ¿¬±¸ |
¿µ¹®Á¦¸ñ(English Title) |
Research for Efficient Massive File I/O on Parallel Programs |
ÀúÀÚ(Author) |
±èÁö¹Î
±è¼±Á¤
È«¼®¹Î
Ji-Min Kim
Sun-Jeong Kim
Seokmin-Hong
Ȳ±ÔÇö
±è¿µÅÂ
Gyuhyeon Hwang
Youngtae Kim
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 18 NO. 02 PP. 0053 ~ 0060 (2017. 04) |
Çѱ۳»¿ë (Korean Abstract) |
ºÐ»ê ¸Þ¸ð¸®ÇüÀÇ º´·Ä ÇÁ·Î±×·¥¿¡¼´Â ÇÁ·Î¼¼¼µéÀÌ µ¶¸³ÀûÀ¸·Î ÀÔÃâ·ÂÀ» ó¸®Çϱ⠶§¹®¿¡ ¿©·¯ À¯ÇüÀÇ ÆÄÀÏ ÀÔÃâ·Â ¹æ½ÄÀÌ »ç¿ëµÈ´Ù. º» ³í¹®¿¡¼´Â ºÐ»ê ¸Þ¸ð¸®Çü º´·Ä ÇÁ·Î±×·¥¿¡¼ÀÇ ´ë¿ë·® ÆÄÀÏ¿¡ ´ëÇÑ È¿À²ÀûÀÎ ÀÔÃâ·Â ¹æ½ÄÀ» ¾Ë¾Æº¸±â À§ÇÏ¿© ´Ù¾çÇÑ ¹æ½ÄÀ» ±¸ÇöÇÏ°í ºñ±³ ºÐ¼®ÇÏ¿´´Ù. ±¸ÇöµÈ ¹æ½ÄÀ¸·Î´Â (i) NFS¸¦ È°¿ëÇÑ º´·Ä ÀÔÃâ·Â ¹æ½Ä, (ii) È£½ºÆ® ÇÁ·Î¼¼¼¿¡¼ÀÇ ¼øÂ÷ ÀÔÃâ·Â°ú µµ¸ÞÀÎ ºÐ»ê ¹æ½Ä, ±×¸®°í (iii) ¸Þ½ÃÁö Àü¼Û Àü¿ë ÀÔÃâ·Â(MPHO) ¹æ½Ä µîÀÌ ÀÖ´Ù. ¼º´É ºÐ¼®À» À§Çؼ º°µµÀÇ ÆÄÀÏ ¼¹ö¸¦ »ç¿ëÇÏ¿´À¸¸ç ÇÑ ´ë ¹× µÎ ´ëÀÇ °è»ê Ŭ¶óÀ̾ðÆ®¿¡¼ ´ÙÁß ÇÁ·Î¼¼¼¸¦ »ç¿ëÇÏ¿´´Ù. ºñ±³ ºÐ¼® °á°ú, ÀÔ·ÂÀÇ °æ¿ì¿¡´Â NFS º´·Ä ÀÔ·Â ¹æ½ÄÀÌ, Ãâ·ÂÀÇ °æ¿ì¿¡´Â µµ¸ÞÀÎ Àü¼ÛÀ» ÅëÇÑ ¼øÂ÷ Ãâ·Â ¹æ½ÄÀÌ °¡Àå È¿À²ÀûÀ¸·Î ³ªÅ¸³µÀ¸¸ç, ¿¹»ó°ú´Â ´Ù¸£°Ô ¸Þ½ÃÁö Àü¼Û Àü¿ë ÀÔÃâ·Â ¹æ½ÄÀÇ ¼º´ÉÀÌ °¡Àå ³·°Ô ³ª¿Ô´Ù. |
¿µ¹®³»¿ë (English Abstract) |
Since processors are handling inputs and outputs independently on distributed memory computers, different file input/output methods are used. In this paper, we implemented and compared various file I/O methods to show their efficiency on distributed memory parallel computers. The implemented I/O systems are as following: (i) parallel I/O using NFS, (ii) sequential I/O on the host processor and domain decomposition, (iii) MPI-IO. For performance analysis, we used a separated file server and multiple processors on one or two computational servers. The results show the file I/O with NFS for inputs and sequential output with domain composition for outputs are best efficient respectively. The MPI-IO result shows unexpectedly the lowest performance. |
Å°¿öµå(Keyword) |
Àΰø Áö´É
À¯Àü ¾Ë°í¸®Áò
ÀýÂ÷Àû ÄÜÅÙÃ÷ »ý¼º
Ç÷¹À̾î ÀûÀÀÇü ¸ó½ºÅÍ »ý¼º
Artificial Intelligence
Genetic Algorithm
Procedural Content Generation
Players Adaptive Monster Generation
º´·Ä ÀÔÃâ·Â
ÁýÇÕ ÀÔÃâ·Â
ºÐ»ê ¸Þ¸ð¸® ÄÄÇ»ÅÍ
MPI-IO
NFS
Parallel I/O
Collective I/O
Distributed memory computer
MPI-IO
NFS
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|