³í¹®¸í |
ÇÖ-½ºÅĹÙÀÌ ±â¹Ý ³Ã°¢½Ã½ºÅÛ ¿îÀüÀ» ÅëÇÑ µ¥ÀÌÅͼ¾ÅÍ ¿î¿µ ½Å·Ú¼º ¹× ¿¡³ÊÁö È¿À²È¿¡ °üÇÑ ¿¬±¸ / Operation Reliability and Energy Optimization for a Hot-standby Cooling System of Data Center |
¼ö·Ï»çÇ× |
´ëÇÑ°ÇÃàÇÐȸ³í¹®Áý, Vol.39 No.3 (2023-03) |
ÆäÀÌÁö |
½ÃÀÛÆäÀÌÁö(189) ÃÑÆäÀÌÁö(11) |
ÁÖÁ¦¾î |
µ¥ÀÌÅͼ¾ÅÍ; ³Ã°¢½Ã½ºÅÛ; ÇÖ-½ºÅĹÙÀÌ; ¿¡³ÊÁö ÃÖÀûÈ; ¿î¿µ ½Å·Ú¼º; PUE(¿¡³ÊÁöÈ¿À²) ; Data Center; Cooling System; Hot-standby; Energy Optimization; Operation Reliability; Power Usage Effectiveness(PUE) |
¿ä¾à1 |
µ¥ÀÌÅͼ¾ÅÍ´Â IT¼ºñ½º º¯È¿¡ µû¸¥ ½Å¼ÓÇÑ ´ëÀÀ»Ó ¾Æ´Ï¶ó ¹«Áß´Ü ¿î¿µÀÇ ¾ÈÁ¤¼º È®º¸¸¦ À§ÇÑ ³Ã°¢½Ã½ºÅÛÀÌ ¸Å¿ì Áß¿äÇÑ ¿ä¼ÒÀÌ´Ù. º» ¿¬±¸´Â ¹«Áß´Ü ¿î¿µ ½Å·Ú¼ºÀ¸·Î ¿ä±¸µÇ´Â µ¥ÀÌÅͼ¾ÅÍÀÇ N+1ÀÌ»ó (¿¹ºñ)ÀåºñÀÇ ÀÌÁßÈ¿Í ¿¡³ÊÁöÈ¿À² ÃÖÀûÈ¿¡ È¿°úÀûÀÎ ´ëÀÀ °¡´ÉÇÑ ÇÖ-½ºÅĹÙÀÌ(HSP)-±â¹Ý ³Ã°¢½Ã½ºÅÛÀ» ÃÖÃÊ·Î Á¦¾ÈÇÏ¿´´Ù. ¶ÇÇÑ PUEcooling ¹× ¿¡³ÊÁöºñ¿ë ´ëÇÑ Á¤·®Àû ºÐ¼®À¸·Î °æÁ¦¼º °üÁ¡¿¡¼µµ °´°üÀûÀÎ ÆǴܱâÁØÀ» Á¦½ÃÇÏ¿´´Ù. µ¥ÀÌÅͼ¾ÅÍÀÇ ³ôÀº ½Å·Ú¼ºÀ» È®º¸¸¦ À§ÇÑ º¸ÆíÀûÀÎ ¹æ¹ýÀ¸·Î N+1ÀÌ»óÀÇ ¿¹ºñ ÀåºñÀÇ ´ë±â ¹æ½ÄÀÌ Ã¤ÅÃµÇ°í ´ë±â ¹æ½Ä¿¡´Â Å©°Ô HSP, WSP ¹× CSP·Î ±¸ºÐµÈ´Ù. ÃÖ±Ù µ¥ÀÌÅͼ¾ÅÍÀÇ ³Ã°¢½Ã½ºÅÛÀÇ ±¸¼ºÀåºñ´Â À¯¿¬ÇÑ IT ºÎÇÏ´ëÀÀÀ» À§ÇØ º¯¼Ó(À¯·®/dz·®)ÀÌ °¡´ÉÇÑ VSD/VFD Àåºñ(³Ãµ¿±â, ÆßÇÁ, ¼Ûdz±â µî)ÀÇ Àû¿ëÀÌ Áõ°¡ÇÏ°í ÀÖÀ¸¸é ÀÏÁ¤¹üÀ§ÀÇ ¿îÀüºÎÇÏ¿¡ ´ëÇÑ ¿¡³ÊÁöÈ¿À²À» ±Ø´ëÈ ÇÒ ¼ö ÀÖ´Ù. Àü¿ë µ¥ÀÌÅͼ¾ÅÍÀÇ Çʼö¿ä°ÇÀÎ N+1 ÀÌ»óÀÇ ³Ã°¢½Ã½ºÅÛ ¿¹ºñÀ²°ú VSD/VFD ÀåºñÀÇ ±¸¼º¿¡ ´ëÇÑ ÃæÁ·Á¶°ÇÀÌ ¸¸Á·µÈ´Ù¸é ¿¡³ÊÁöÈ¿À² Ãø¸é¿¡¼ HSP-±â¹Ý ³Ã°¢½Ã½ºÅÛ Àû¿ëÀ» ¿ì¼± °í·ÁÇØ¾ß ÇÑ´Ù. 30MW±Þ ´ë±Ô¸ð Aµ¥ÀÌÅͼ¾ÅÍÀÇ »ç·ÊºÐ¼®À» ÅëÇÑ CSP¿Í HSP ¿îÀüÈ¿°ú¸¦ ºñ±³ÇÑ °á°ú, µ¥ÀÌÅͼ¾ÅÍ ÀüüÀÇ ³Ã°¢½Ã½ºÅÛ ¼ÒºñÀü·ÂÀº µ¿ÀÏÇÑ ¼³°è¿ë·®À» ±âÁØÀ¸·Î HSP¸¦ Àû¿ëÇÏ´Â °ÍÀÌ CSP ´ëºñ ¾à 15% °¨¼ÒÇÑ´Ù. ±âÁ¸ÀÇ CSP-±â¹Ý ³Ã°¢½Ã½ºÅÛÀÇ PUEcoolingÀº 1.23ÀÌ°í ³Ã°¢ ¿¡³ÊÁöºñ¿ëÀº 193 USD/IT-kW°¡ ¼Ò¿äµÈ´Ù. HSP-±â¹Ý ³Ã°¢½Ã½ºÅÛÀÇ PUEcoolingÀº 1.20ÀÌ°í ³Ã°¢ ¿¡³ÊÁöºñ¿ëÀº 163 USD/IT-kW°¡ ¼Ò¿äµÇ¾î ¿¡³ÊÁöÈ¿À² °³¼±ÀÇ È¿°ú°¡ ÃæºÐÇÏ´Ù. Å« ½Ã¼³º¯°æ ¹× ÅõÀÚ ¾øÀÌ ³Ã°¢¿¡³ÊÁö¸¦ È¿°úÀûÀ¸·Î Àý°¨ÇÒ ¼ö ÀÖ´Â ÃÖ¼±ÀÇ Àü·«Àº HSP-±â¹Ý ³Ã°¢½Ã½ºÅÛÀÌ´Ù. |
¿ä¾à2 |
In a data center, a cooling system is a crucial factor in ensuring uninterrupted operation and quick response to changes in IT services. This study proposes a cooling system based on hot standby sparing (HSP) that can effectively provide redundancy and optimize energy efficiency in data centers with N+1 or more equipment required for reliable uninterrupted operation. The standby strategy adopted for ensuring high reliability in data centers was classified into HSP, warm standby (WSP), and cold standby (CSP). If the cooling system meets the redundancy requirement of N+1 or higher, which is necessary for a dedicated data center, and the configuration of VSD/VFD equipment, then the HSP-based cooling system should be considered first in terms of energy efficiency. Through a case analysis of 30 MW-class large-scale A-data centers, the CSP-based and HSP-based cooling systems were compared. The results showed that the cooling power consumption for the same design capacity was reduced by approximately 15% when applying the HSP-based system compared to the CSP system. The power usage effectiveness (PUE) of the HSP-based cooling system was 1.20, and the cooling energy cost was 163 USD/IT-kW, indicating that the HSP-based system significantly improved energy efficiency. Thus, the HSP-based cooling system is the best strategy for effectively reducing cooling energy without requiring major facility changes or investments. |