• 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 :

ÇѱÛÁ¦¸ñ(Korean Title) ±â°èÇнÀ ±â¹Ý FaaS Ŭ¶ó¿ìµå ¸®Àü ¼±Åà ±â¹ý
¿µ¹®Á¦¸ñ(English Title) Machine Learning Based FaaS Cloud Region Selection Method
ÀúÀÚ(Author) Á¶¼º¹è   ÇѱâÁØ   ±èµ¿±Õ   Sungbae Jo   Kijun Han   Dongkyun Kim  
¿ø¹®¼ö·Ïó(Citation) VOL 27 NO. 07 PP. 0325 ~ 0330 (2021. 07)
Çѱ۳»¿ë
(Korean Abstract)
FaaS(Function as a Service)´Â ºÐ»êµÈ Ŭ¶ó¿ìµå ÀÚ¿ø¿¡ ÇÔ¼ö¸¦ µî·ÏÇÏ°í ½ÇÇàµÇ´Â Ƚ¼ö¸¸Å­ ºñ¿ëÀ» ÁöºÒÇÏ´Â ¹æ½ÄÀÇ Å¬¶ó¿ìµå ¼­ºñ½º¸¦ ¸»Çϸç ÀÌ´Â ÁÖ·Î ¼­¹ö¸®½º(Serverless) ÄÄÇ»ÆÃÀ» ±¸ÇöÇϱâ À§ÇØ »ç¿ëµÈ´Ù. ÁÖ¿ä Ŭ¶ó¿ìµå °ø±Þ¾÷ü°¡ ÇöÀç ¼­¹ö¸®½º ÄÄÇ»Æà Ç÷§ÆûÀ» Á¦°øÇÔ¿¡ µû¶ó ¸¹Àº ¼­ºñ½º°¡ FaaS±â¹ÝÀ¸·Î ±¸ÃàµÇ°í ÀÖ´Ù. °³º° ÇÔ¼ö´Â Àü ¼¼°è ¿©·¯ Áö¿ª¿¡ Èð¾îÁ®ÀÖ´Â º¹¼öÀÇ Å¬¶ó¿ìµå ¸®Àü(Region)¿¡ ¹èÆ÷ÇÒ ¼ö ÀÖÀ¸¸ç, ÀϹÝÀûÀ¸·Î ´Ü¼øÈ÷ Áö¸®ÀûÀ¸·Î °¡Àå °¡±î¿î ¸®ÀüÀÇ ÇÔ¼ö¸¦ ¼±ÅÃÇÏ¿© »ç¿ëÇÏ°Ô µÈ´Ù. ÇÏÁö¸¸ ·Î±×¸¦ ºÐ¼®Çغ¸¸é Áö¸®ÀûÀ¸·Î °¡Àå °¡±î¿î ¸®ÀüÀÇ ÀÀ´ä¼Óµµ°¡ Ç×»ó °¡Àå ºü¸¥ °ÍÀÌ ¾Æ´Ï¶ó´Â °ÍÀ» ¾Ë ¼ö ÀÖ°í, º¸´Ù ºü¸¥ ¸®ÀüÀ» ¿¹ÃøÇÏ¿© ¼±ÅÃÇÒ ¼ö ÀÖ´Ù¸é Ãß°¡ÀûÀÎ ¼­ºñ½º ÀÀ´ä ¼Óµµ °³¼±ÀÌ °¡´ÉÇÏ´Ù. º» ¿¬±¸¿¡¼­´Â FaaSÀÇ ÀÀ´ä ½Ã°£À» ±âÁØÀ¸·Î °¢ ¸®ÀüÀÇ ·Î±×¸¦ ºñ±³¡¤ºÐ¼®ÇÏ°í ±â°èÇнÀÀ» ÅëÇØ ÃÖÀûÀÇ Áö¿ªÀ» ¿¹ÃøÇÏ´Â ¹æ¹ýÀ» Á¦½ÃÇÏ°íÀÚ ÇÑ´Ù.
¿µ¹®³»¿ë
(English Abstract)
FaaS (Function as a Service) is a cloud service that registers functions on distributed cloud resources and remunerates these cloud resources each time they run. FaaS is mainly used to implement serverless computing. As major cloud providers currently provide serverless computing platforms, many services are being built on the basis of FaaS. Individual functions can be deployed in multiple cloud regions scattered across multiple regions of the world, and in general, simply select and use the function in the geographically closest region. However, by analyzing the log, it can be seen that the response speed of the geographically closest region is not always the fastest, and service response speed could be improved if a faster region could be predicted and selected in advance. In this study, we compared and analyzed the log of each region based on the response time of FaaS and proposed a method to predict the optimal region through machine learning.
Å°¿öµå(Keyword) FaaS   ¼­¹ö¸®½º   ¼­¹ö¸®½º ÄÄÇ»Æà  Ŭ¶ó¿ìµå ¸®Àü   ±â°è ÇнÀ   XGBoost   Function as a Service (FaaS)   serverless computing   cloud region   machine learning   XGBoost  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå