µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
CUBRID RDBMS »ó¿¡¼ º´·Ä ÁúÀÇ Ã³¸®¸¦ Áö¿øÇÏ´Â ºÐ»ê ¹Ìµé¿þ¾î °³¹ß |
¿µ¹®Á¦¸ñ(English Title) |
Development of Distributed Middleware supporting Parallel Query Processing on CUBRID RDBMS |
ÀúÀÚ(Author) |
ÃÖ¹®Ã¶
Á¶¾Æ¶ó
À±¹Î
±èÇüÀÏ
ÀåÀç¿ì
Mun-Chul Choi
Ahra Cho
Min Yoon
Hyeong-Il Kim
Jae-Woo Chang
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 30 NO. 02 PP. 0067 ~ 0077 (2014. 08) |
Çѱ۳»¿ë (Korean Abstract) |
ÃÖ±Ù ±Þ¼Óµµ·Î Áõ°¡ÇÏ´Â ÀÎÅÍ³Ý »ç¿ëÀÚ ¹× SNSÀÇ ±Þ°ÝÇÑ È®»êÀ¸·Î Á¤º¸ÀÇ ¾çÀÌ ±âÇϱ޼öÀûÀ¸·Î Áõ°¡ÇÏ¿´À¸¸ç, µû¶ó¼ ºòµ¥ÀÌÅÍ¿¡ ´ëÇÑ ¿¬±¸°¡ È°¹ßÈ÷ ÀÌ·ç¾îÁö°í ÀÖ´Ù. ºòµ¥ÀÌÅ͸¦ ´Ù·ç´Â NoSQL¿¡ ´ëÇÑ ¿¬±¸°¡ È°¹ßÈ÷ ÁøÇàµÇ°í ÀÖÁö¸¸, »ç¿ëÀÚ ÆíÀǼº°ú µ¥ÀÌÅͺ£À̽ºÀÇ ACID Á¶°ÇÀ» ¸¸Á·ÇÏÁö ¸øÇÏ´Â ¹®Á¦Á¡ÀÌ Á¸ÀçÇÑ´Ù. µû¶ó¼ RDBMS¸¦ ±â¹ÝÀ¸·Î ºòµ¥ÀÌÅÍ Ã³¸®¸¦ ¼öÇàÇÏ´Â ¿òÁ÷ÀÓÀÌ È°¹ßÇØÁö°í ÀÖ´Ù. À̸¦ À§ÇÑ ´ëÇ¥ÀûÀÎ ±â¹ýÀÎ CUBRID SHARD´Â µ¥ÀÌÅͺ£À̽º¸¦ ¼öÆò ºÐÇÒÇÏ¿© °¢±â ´Ù¸¥ ¹°¸® ³ëµå¿¡ Shard ´ÜÀ§·Î µ¥ÀÌÅ͸¦ ³ª´©¾î¼ ÀúÀåÇÏ¿©, µ¥ÀÌÅÍÀÇ ºÐ»ê ÀúÀåÀ» Áö¿øÇÑ´Ù. ±×·¯³ª ÇØ´ç ±â¹ýÀº ÁúÀÇÀÇ ¿Ïº®ÇÑ º´·Ä 󸮰¡ ºÒ°¡´ÉÇϱ⠶§¹®¿¡, ÇÑ Å¬¶óÀ̾ðÆ®ÀÇ µ¥ÀÌÅͺ£À̽º°¡ ´Ù¼öÀÇ ¼¹ö¿¡ ºÐ»ê ÀúÀåµÇ¾î ÀÖ´Â °æ¿ì ÁúÀÇ Áý°è µî ´Ù¼öÀÇ ¼¹ö¿¡¼ ÁúÀÇ Ã³¸®¸¦ ¼öÇàÇÏÁö ¸øÇÏ´Â ¹®Á¦Á¡ÀÌ Á¸ÀçÇÑ´Ù. µû¶ó¼ º» ³í¹®¿¡¼´Â ºÐ»ê µ¥ÀÌÅÍ Ã³¸® ȯ°æ¿¡¼ º´·Ä ÁúÀÇ Ã³¸®»Ó¸¸ ¾Æ´Ï¶ó ´Ù¾çÇÑ Áý°èÁúÀÇ Ã³¸®¸¦ Áö¿øÇÏ´Â CUBRID ±â¹Ý ºÐ»ê ¹Ìµé¿þ¾î¸¦ Á¦¾ÈÇÑ´Ù. |
¿µ¹®³»¿ë (English Abstract) |
The wide spread of Internet services and SNS (Social Network Service) has produced a hugh volume of data, thus researches dealing with big data has gained significant attentions. NoSQL is famous for big data processing since it allows agile processing of information on a massive scale. However, it has limitations that it does not satisfy ACID condition of database system and it provides inefficient usability. Therefore, RDBMS has been spotlighted as a new wave of big data processing. CUBRID SHARD is designed to provide distributed load balancing by allowing unlimited number of database shards stored in physical nodes. However, CUBRID does not support efficient query processing over distributed data, so aggregate queries cannot be performed on the existing CUBRID. To solve this problem, we CUBRID-based distributed middleware that supports not only parallel query processing, but also various aggregation query processing in distributed data processing environments. |
Å°¿öµå(Keyword) |
RDBMS
ºòµ¥ÀÌÅÍ
ºÐ»ê ó¸®
º´ÇÕ Ã³¸® ½Ã½ºÅÛ
CUBRID
RDBMS
big data
distributed processing
merge processing system
CUBRID
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|