• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹Á¤º¸°úÇÐȸ ³í¹®Áö > Á¤º¸°úÇÐȸ ÄÄÇ»ÆÃÀÇ ½ÇÁ¦ ³í¹®Áö (KIISE Transactions on Computing Practices)

Á¤º¸°úÇÐȸ ÄÄÇ»ÆÃÀÇ ½ÇÁ¦ ³í¹®Áö (KIISE Transactions on Computing Practices)

Current Result Document : 4 / 6 ÀÌÀü°Ç ÀÌÀü°Ç   ´ÙÀ½°Ç ´ÙÀ½°Ç

ÇѱÛÁ¦¸ñ(Korean Title) ºòµ¥ÀÌÅÍ Ç÷§Æû¿¡¼­ ÀÌÁ¾ ¼­ºñ½º°£ ¼º´É °£¼· Çö»ó Á¦¾î¿¡ °üÇÑ ¿¬±¸
¿µ¹®Á¦¸ñ(English Title) Research of Performance Interference Control Technique for Heterogeneous Services in Bigdata Platform
ÀúÀÚ(Author) Áø±â¼º   ÀÌ»ó¹Î   ±è¿µ±Õ   Kisung Jin   Sangmin Lee   Youngkyun Kim  
¿ø¹®¼ö·Ïó(Citation) VOL 22 NO. 06 PP. 0284 ~ 0289 (2016. 06)
Çѱ۳»¿ë
(Korean Abstract)
Hadoop ±â¹ÝÀÇ ºòµ¥ÀÌÅÍ ºÐ¼® ¸ðµ¨¿¡¼­´Â ¿ø½Ã µ¥ÀÌÅ͸¦ »ý»êÇÏ´Â ÀÀ¿ë°è ½Ã½ºÅÛ°ú À̸¦ ºÐ¼®Çϱâ À§ÇÑ ºÐ¼®°è ½Ã½ºÅÛ°£ÀÇ µ¥ÀÌÅÍ À̵¿ÀÌ ºÒ°¡ÇÇÇÏ´Ù. ÀÌ¿¡ µû¶ó, ÀÀ¿ë ¼­ºñ½º¿Í ºÐ¼® ¼­ºñ½º¸¦ ÇϳªÀÇ Ç÷§Æû¿¡¼­ µ¿½Ã¿¡ Áö¿øÇÒ ¼ö ÀÖ´Â À¯´ÏÆÄÀÌµå ºòµ¥ÀÌÅÍ ÆÄÀϽýºÅÛ ±â¼úÀÌ ¼Ò°³µÇ°í ÀÖ´Ù. ±×·¯³ª, ´ÜÀÏ Ç÷¡Æû ¿î¿µ¿¡ µû¸¥ °æÁ¦¼º, ÀÚ¿ø È¿À²¼º µî ´Ù¾çÇÑ Ãø¸é¿¡¼­ÀÇ ÀåÁ¡¿¡µµ ºÒ±¸ÇÏ°í ÇöÀç ±â¼ú ¼öÁØ¿¡¼­´Â ÀÀ¿ë ¼­ºñ½º¿Í ºÐ¼® ¼­ºñ½ºÀÇ »óÈ£ °£¼·¿¡ ÀÇÇÑ ¼º´É ÀúÇÏ Çö»óÀ» ±Øº¹ÇÏ´Â °ÍÀÌ °¡Àå Å« ´ç¸é °úÁ¦·Î ³²¾Æ ÀÖ´Ù. º» ³í¹®¿¡¼­´Â À̸¦ ÇØ°áÇϱâ À§ÇÑ ÀÏÂ÷Àû ´Ü°è·Î µÎ ¼­ºñ½º¿¡ ´ëÇØ ½Ç¼­ºñ½º ¼öÁØ ½Ã¹Ä·¹À̼ÇÀ» ÅëÇØ ½Ã½ºÅÛ ÀÚ¿øÀÇ È°¿ë·ü, ¿öÅ©·Îµå Ư¼º, ÀÔÃâ·Â ºÒ±ÕÇüÀÇ ¼¼ °¡Áö °üÁ¡¿¡¼­ °üÂûÇÑ ÈÄ ¼º´É °£¼· ¹®Á¦ÀÇ ±Ùº»ÀûÀÎ ¿øÀÎÀ» µµÃâÇÏ¿´´Ù. ¶ÇÇÑ À̸¦ ÇØ°áÇϱâ À§ÇÑ ¹æ¹ýÀ¸·Î ù°, µ¥ÀÌÅÍ ¼­¹öÀÇ ÀÔÃâ·Â °æ·Î¸¦ ºÐ¸®ÇÏ¿© ÀÀ¿ë ¼­ºñ½º¿Í ºÐ¼® ¼­ºñ½º °¢°¢ µ¶¸³ÀûÀÎ ÀÔÃâ·Â °èÃþÀ» ±¸¼ºÇÏ´Â ±¸Á¶ÀûÀÎ ÇØ°áÃ¥°ú, µÑ°, ¼øÂ÷ Àбâ Ư¼ºÀ» °¡Áö´Â ºÐ¼® ¼­ºñ½º ÀÔÃâ·Â Ư¼ºÀÇ È¿°ú¸¦ ±Ø´ëÈ­Çϱâ À§ÇÑ ¼±Á¦Àû ¹Ì¸® Àб⠱â¹ýÀÇ ±â¼úÀû ÇØ°áÃ¥À» Á¦¾ÈÇÑ´Ù. ÇÑÆí, ³í¹®¿¡¼­ Á¦¾ÈÇÑ ¹æ¹ýÀÇ È¿°ú¸¦ °ËÁõÇϱâ À§ÇØ ½Ã¹Ä·¹À̼ǰú µ¿ÀÏÇÑ ¹æ¹ýÀÇ ½ÃÇèÀ» ±âÁ¸ ½Ã½ºÅÛ°ú Á¦¾ÈÇÑ ½Ã½ºÅÛ °¢°¢¿¡ ´ëÇØ ¼öÇàÇÑ °á°ú ±âÁ¸ ½Ã½ºÅÛ ´ëºñ ¿ì¼öÇÑ ¼º´ÉÀ» È®ÀÎÇÒ ¼ö ÀÖ¾ú´Ù.
¿µ¹®³»¿ë
(English Abstract)
In the Hadoop-based Big Data analysis model, the data movement between the legacy system and the analysis system is difficult to avoid. To overcome this problem, a unified Big Data file system is introduced so that a unified platform can support the legacy service as well as the analysis service. However, major challenges in avoiding the performance degradation problem due to the interference of two services remain. In order to solve this problem, we first performed a real-life simulation and observed resource utilization, workload characteristics and I/O balanced level. Based on this analysis, two solutions were proposed both for the system level and for the technical level. In the system level, we divide I/O path into the legacy I/O path and the analysis I/O path. In the technical level, we introduce an aggressive prefetch method for analysis service which requires the sequential read. Also, we introduce experimental results that shows the outstanding performance gain comparing the previous system.
Å°¿öµå(Keyword) ºòµ¥ÀÌÅÍ   ÆÄÀϽýºÅÛ   À¯´ÏÆÄÀÌµå ºòµ¥ÀÌÅÍ ÆÄÀϽýºÅÛ   ÀÔÃâ·Â °£¼·   bigdata   filesystem   unified bigdata filesystem   I/O interference  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå