Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)
ÇѱÛÁ¦¸ñ(Korean Title) |
´ë±Ô¸ð °úÇÐÀÀ¿ëÀ» À§ÇÑ È¿À²ÀûÀÎ ºÐ»ê ½´ÆÛÄÄÇ»Æà ÀÚ¿ø°ü¸® ±â¼ú ¿¬±¸ |
¿µ¹®Á¦¸ñ(English Title) |
Effective Distributed Supercomputing Resource Management for Large Scale Scientific Applications |
ÀúÀÚ(Author) |
³ë½Â¿ì
±èÁ÷¼ö
±è»ó¿Ï
±è¼¿µ
Ȳ¼ø¿í
Seungwoo Rho
Jik-Soo Kim
Sangwan Kim
Seoyoung Kim
Soonwook Hwang
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 42 NO. 05 PP. 0573 ~ 0579 (2015. 05) |
Çѱ۳»¿ë (Korean Abstract) |
±¹°¡ ½´ÆÛÄÄÇ»Æà ÀÎÇÁ¶ó´Â ±¹³» ¿©·¯ Áö¿ª¿¡ ºÐ»êµÈ ½´ÆÛÄÄÇ»Æà Ŭ·¯½ºÅ͵é·Î ÀÌ·ç¾îÁ® ÀÖÀ¸¸ç, º» ¿¬±¸ÆÀ¿¡¼´Â ÀÌ·¯ÇÑ À̱âÁ¾ÀÇ Áö¸®ÀûÀ¸·Î ºÐ»êµÈ Ŭ·¯½ºÅ͵éÀ» ´ë±Ô¸ð °úÇÐ ÀÀ¿ë ¿¬±¸Àڵ鿡°Ô È¿À²ÀûÀ¸·Î Á¦°øÇϱâ À§ÇØ ´ë±Ô¸ð °è»êó¸® ½Ã½ºÅÛÀÎ HTCaaS(High-Throughput Computing as a Service)¸¦ ÀÚü °³¹ßÇÏ¿´´Ù. º» ³í¹®¿¡¼´Â ÀÌ·¯ÇÑ ´ë±Ô¸ð °è»êó¸® ½Ã½ºÅÛ(HTCaaS)À» È°¿ëÇÏ¿© °¢ °è»ê ÀÚ¿øÀ» µ¿ÀûÀ¸·Î °ü¸®ÇÏ´Â ¹æ¹ý¿¡ ´ëÇؼ ³íÀÇÇÏ°í, ½Å¾àÀçâÃâÀ̶ó´Â ½ÇÁ¦ °úÇÐ ÀÀ¿ëÀ» ÅëÇØ ±× È¿À²¼ºÀ» °ËÁõÇÑ´Ù. ƯÈ÷ À¯È¿ ÀÚ¿ø ½Äº°À» À§ÇÑ ´ë±â½Ã°£ ¹× ¼º°ø·ü °³³äÀ» ÀÌ¿ëÇÑ µ¿Àû °è»ê ÀÚ¿ø °ü¸® ±â¼úÀ» Àû¿ëÇÔÀ¸·Î½á ÀÚ¿ø È°¿ë·ü°ú Á¤È®¼º, ½Å·Ú¼º, ÆíÀǼºÀÌ Çâ»óµÉ ¼ö ÀÖÀ¸¸ç, ±× °á°ú ÀüüÀûÀÎ ÀÛ¾÷ ½Ã°£ÀÇ ´ÜÃà°ú ÀÛ¾÷ 󸮷®µµ Çâ»óµÉ ¼ö ÀÖÀ½À» È®ÀÎÇÒ ¼ö ÀÖ¾ú´Ù.
|
¿µ¹®³»¿ë (English Abstract) |
Nationwide supercomputing infrastructures in Korea consist of geographically distributed supercomputing clusters. We developed High-Throughput Computing as a Service(HTCaaS) based on these distributed national supecomputing clusters to facilitate the ease at which scientists can explore large-scale and complex scientific problems. In this paper, we present our mechanism for dynamically managing computing resources and show its effectiveness through a case study of a real scientific application called drug repositioning. Specifically, we show that the resource utilization, accuracy, reliability, and usability can be improved by applying our resource management mechanism. The mechanism is based on the concepts of waiting time and success rate in order to identify valid computing resources. The results show a reduction in the total job completion time and improvement of the overall system throughput.
|
Å°¿öµå(Keyword) |
HTCaaS
½´ÆÛÄÄÇ»ÅÍ
HTC
MTC
µ¿Àû °è»ê ÀÚ¿ø °ü¸®
HTCaaS
supercomputer
HTC
MTC
dynamic compute resource management
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|