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

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ÇÐȸÁö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ÇÐȸÁö > µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

Current Result Document : 11 / 11

ÇѱÛÁ¦¸ñ(Korean Title) ºÐ»ê ½ºÆ®¸² ÀÀ¿ëÀÇ º¥Ä¡¸¶Å© Å×½ºÆ®¸¦ À§ÇÑ Apache StormÀÇ ¼º´É ÃøÁ¤ ±â¹ý
¿µ¹®Á¦¸ñ(English Title) Performance Measurement Techniques of Apache Storm for Benchmark Test of Distributed Stream Applications
ÀúÀÚ(Author) ¼Õ½Ã¿î   ¹®¾ç¼¼   Siwoon Son   Yang-Sae Moon  
¿ø¹®¼ö·Ïó(Citation) VOL 36 NO. 03 PP. 0105 ~ 0115 (2020. 12)
Çѱ۳»¿ë
(Korean Abstract)
ÃÖ±Ù, ¼Ò¼È ³×Æ®¿öÅ· ¼­ºñ½º(SNS), »ç¹° ÀÎÅͳÝ(IoT) µî¿¡¼­ ´ë¿ë·®À¸·Î ¹ß»ýÇÏ´Â µ¥ÀÌÅÍ ½ºÆ®¸²À» ½Ç½Ã °£À¸·Î ºü¸£°Ô ó¸®Çϱâ À§ÇØ µ¥ÀÌÅÍ ½ºÆ®¸² ó¸® ±â¼ú ¿¬±¸°¡ È°¹ßÇÏ´Ù. ƯÈ÷, ºÐ»ê ȯ°æ¿¡¼­ ³ôÀº ó¸® ¼º ´ÉÀ» º¸ÀÌ´Â ºÐ»ê ½ºÆ®¸² ó¸® ¿£ÁøÀÎ Apache StormÀº ±âÁ¸ ½ºÆ®¸² ÀÀ¿ëÀ» È®ÀåÇÑ ºÐ»ê ½ºÆ®¸² ÀÀ¿ë °³¹ß ¿¡ ¸¹ÀÌ È°¿ëµÇ°í ÀÖ´Ù. ÀÌ·¯ÇÑ ÀÀ¿ëÀÇ ºÐ»êÈ­¸¦ ÅëÇÑ ¼º´É Çâ»óÀ» Á¤·®ÀûÀ¸·Î Æò°¡Çϱâ À§ÇØ ºÐ»ê ½ºÆ®¸² ÀÀ¿ëÀÇ º¥Ä¡¸¶Å© Å×½ºÆ®´Â ¸Å¿ì Áß¿äÇÏ´Ù. ÇÏÁö¸¸, ºÐ»ê ȯ°æ¿¡ ¹èÄ¡µÈ ´Ù¼öÀÇ Å½ºÅ©·ÎºÎÅÍ ºÐ»ê ½ºÆ®¸² ÀÀ ¿ëÀÇ Ã³¸® ¼º´ÉÀ» ÃßÀûÇÏ°í ¼öÁýÇÏ´Â °ÍÀº º¹ÀâÇÑ ÀÛ¾÷ÀÌ´Ù. º» ³í¹®¿¡¼­´Â Storm¿¡¼­ µ¿ÀÛÇÏ´Â ºÐ»ê ½ºÆ®¸² ÀÀ¿ëÀÇ Ã³¸® ¼º´ÉÀ» ÃøÁ¤ÇÏ´Â ´Ù¼¸ °¡Áö º¥Ä¡¸¶Å© ±â¹ýµéÀ» ¼³¸íÇÏ°í À̵éÀÇ Àå´ÜÁ¡À» ºÐ¼®ÇÑ´Ù. °¢°¢Àº ·Î Äà ÆÄÀÏ ½Ã½ºÅÛ ±â¹Ý, µ¥ÀÌÅͺ£À̽º ±â¹Ý, ¸Þ½ÃÁö Å¥ ±â¹Ý, REST API ±â¹Ý, ¸ÞÆ®¸¯ ¼­¹ö ±â¹ÝÀ̸ç, À̵éÀÇ ºñ±³¸¦ ÅëÇØ ºÐ»ê ½ºÆ®¸² ÀÀ¿ëÀÇ Æ¯¼º¿¡ µû¶ó ÀûÇÕÇÑ º¥Ä¡¸¶Å© Å×½ºÆ®¸¦ ¼öÇàÇÒ ¼ö ÀÖ´Ù.
¿µ¹®³»¿ë
(English Abstract)
In recent years, researches on data stream processing technology have been actively conducted in order to rapidly process a large amount of data streams generated in real-time in social networking services (SNS) and Internet of Things (IoT). In particular, Apache Storm, a distributed stream processing engine showing high processing performance in a distributed environment, is widely used for developing distributed stream applications that extend existing stream applications. Benchmark tests for distributed stream applications are very important to quantitatively evaluate the performance improvement through decentralization of these applications. However, it is a complex task to track and collect the processing performance of distributed stream applications from multiple tasks deployed in a distributed environment. In this paper, we describe five benchmark test techniques that measure the processing performance of distributed stream applications working in Storm, and analyze their strengths and weaknesses. Each utilizes a local file system, database, message queue, REST API, and metric server, and through their comparison, we can perform appropriate benchmark tests according to the characteristics of distributed stream applications.
Å°¿öµå(Keyword) ºÐ»ê 󸮠  ½Ç½Ã°£ 󸮠  µ¥ÀÌÅÍ ½ºÆ®¸²   º¥Ä¡¸¶Å© Å×½ºÆ®   Distributed processing   real-time processing   data stream   benchmark test  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå