Á¤º¸°úÇÐȸ ³í¹®Áö D : µ¥ÀÌŸº£À̽º
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
°¡»óÈµÈ Hadoop Ŭ·¯½ºÅÍ È¯°æ¿¡¼ VM°ú ŽºÅ©ÀÇ ºÐ¼® ¹× ¹èÄ¡ |
¿µ¹®Á¦¸ñ(English Title) |
Placement and Analysis of VM and Task in Virtualized Hadoop Cluster Environments |
ÀúÀÚ(Author) |
±èÅ¿ø
±èÇöÁØ
±èÁظð
Tae-Won Kim
HyunJun Kim
Joon-Mo Kim
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 40 NO. 05 PP. 0289 ~ 0294 (2013. 10) |
Çѱ۳»¿ë (Korean Abstract) |
³×Æ®¿öÅ©ÀÇ ¹ßÀü°ú ÀÎÅÍ³Ý È®»ê¿¡ ÈûÀÔ¾î ¼ºñ½º µ¥ÀÌÅÍ°¡ °íÇ°Áú, ´ë¿ë·®È µÇ¾ú´Ù. µû¶ó¼ ºÐ»ê ó¸®ÀÇ Çʿ伺ÀÌ ´ëµÎµÇ°í ¿¬±¸°¡ È°¹ßÈ÷ ÁøÇàµÇ°í ÀÖ´Ù. ¶ÇÇÑ, ¼¹öÀÇ ½Ã½ºÅÛ ÀÚ¿øÀ» È¿À²ÀûÀ¸·Î Á¦°øÇÒ ¼ö ÀÖ´Â ¼ºñ½º °¡»óÈ ±â¼úÀÌ ÁÖ¸ñ¹Þ°í ÀÖ´Ù. ±×·¯³ª °¡»ó ¸Ó½Å ȯ°æ¿¡¼ÀÇ ´ë¿ë·® µ¥ÀÌÅÍ ºÐ»êó¸® ½Ã½ºÅÛÀ» ±¸¼ºÇÒ °æ¿ì ¸¹Àº ¹®Á¦°¡ ¹ß»ýÇÏ°Ô µÈ´Ù. º» ³í¹®¿¡¼´Â °¡»ó ¸Ó½Å ȯ°æ¿¡¼ÀÇ °¡»ó ¸Ó½Å°ú ŽºÅ©µéÀÇ »ý¼º ¹èÄ¡¿¡ µû¸¥ I/O ´ë¿ªÆøÀÇ ÃÖÀûÈ ½ÇÇèÀ» ÇÏ°í, °á°ú¸¦ Æò°¡ÇÑ´Ù. ¶ÇÇÑ º» ³í¹®¿¡¼ ¼öÇàÇÑ ½ÇÇè °á°ú´Â °¡»ó ¸Ó½Å ȯ°æ¿¡¼ÀÇ I/O ´ë¿ªÆø ¹ë·±½Ì(balancing)À» Áö¿øÇÏ´Â ÇÏµÓ ½ºÄÉÁì·¯(Hadoop Scheduler)ÀÇ ¿¬±¸¿¡ »ç¿ëµÉ ¿¹Á¤ÀÌ´Ù. |
¿µ¹®³»¿ë (English Abstract) |
Recently, distributed processing systems for big data have been actively investigated owing to the development of high speed network and storage technologies. In addition, a virtual system that can provide efficient use of system resources through the consolidation of servers has been increasingly recognized. But, when we configure a distributed processing system for big data in virtual machine environments, many problems occur. In this paper, we did an experiment on the optimization of I/O bandwidth according to the creation and placement of VMs and tasks with composing a Hadoop cluster in virtual environments and evaluated the results of the experiment. These results conducted by this paper will be used in the study on the development of a Hadoop Scheduler supporting I/O bandwidth balancing in virtual environments. |
Å°¿öµå(Keyword) |
ÇϵÓ
ºÐ»ê µ¥ÀÌÅÍ º£À̽º
°¡»ó ¸Ó½Å
ºò µ¥ÀÌÅÍ
hadoop
distributed database
virtual machine
big data
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|