ÀÚ·á°Ë»ö-Ç¥ÁØ

Ȩ > ÀڷḶ´ç > ÀÚ·á°Ë»ö > Ç¥ÁØ

ÀÚ·á °Ë»ö°á°ú

°Ë»öÆäÀÌÁö·Î
Ç¥ÁØÁ¾·ù Á¤º¸Åë½Å´ÜüǥÁØ(TTAS)
Ç¥ÁعøÈ£ TTAK.KO-10.1630 ±¸ Ç¥ÁعøÈ£
Á¦°³Á¤ÀÏ 2025-12-05 ÃÑ ÆäÀÌÁö 15
ÇÑ±Û Ç¥ÁØ¸í ¸Þ¸ð¸® Á᫐ ÄÄÇ»ÆÃ - ±â´É ¿ä±¸»çÇ×
¿µ¹® Ç¥Áظí Memory Centric Computing - Functional Requirements
ÇÑ±Û ³»¿ë¿ä¾à º» Ç¥ÁØ¿¡¼­´Â ¸Þ¸ð¸® Á᫐ ÄÄÇ»ÆÃÀ» È®Àå ¸Þ¸ð¸®¸¦ ¿ÜºÎ¿¡ °®°í ³»ºÎ ¸Þ¸ð¸®¿Í´Â ´Ù¸¥ º°µµÀÇ ¸Þ¸ð¸® ¿¬°á¸ÁÀ¸·Î ¿ÜºÎ È®Àå ¸Þ¸ð¸®¸¦ Á¢±ÙÇÏ´Â °³³äÀÇ ÄÄÇ»ÆÃÀ¸·Î ¼³¸íÇÑ´Ù. ±âÁ¸ ÄÄÇ»ÆÃ ½Ã½ºÅÛ°ú´Â ´Þ¸®, ÀÌ´õ³Ý ±â¹ÝÀÇ ³×Æ®¿öÅ©¸¦ »ç¿ëÇÏÁö ¾Ê°í È®Àå ¸Þ¸ð¸®¸¦ Á¢±ÙÇϹǷΠÁö¿¬ ½Ã°£ ¹× °øÀ¯ °üÁ¡¿¡¼­ ±¸Á¶»ó À¯¸®ÇÏ´Ù. ¾Æ¿ï·¯, º» Ç¥ÁØ¿¡¼­´Â ¸Þ¸ð¸® Á᫐ ÄÄÇ»ÆÃÀÇ ÇÙ½É ±â¼úÀÎ ¸Þ¸ð¸® ¿¬°á¸ÁÀÇ ±â´É ¹× ±¸Á¶¸¦ ¼³¸íÇϰí, ¸Þ¸ð¸® Á᫐ ÄÄÇ»ÆÃÀ» Àû¿ëÇÏ¿© È®ÀåÇÒ ¼ö ÀÖ´Â ·¢ ÄÄÇ»ÆÃ, ¸Þ¸ð¸® °øÀ¯ ÄÄÇ»ÆÃ, ÀÚ¿ø ºÐ¸® ÄÄÇ»ÆÃ ¸ðµ¨ÀÇ ±â´É¿Í ±â´É ¿ä±¸»çÇ×À» Á¦½ÃÇÑ´Ù.
¿µ¹® ³»¿ë¿ä¾à This standard defines memory-centric computing as a paradigm where extended memory resides externally and is accessed via a dedicated memory interconnect, distinct from internal memory. Unlike conventional computing systems, which typically rely on Ethernet-based networks for accessing extended memory, this approach offers architectural advantages in terms of latency and sharing. Furthermore, this standard elucidates the functionalities and architecture of the memory interconnect, a pivotal technology in memory-centric computing. It also presents the functionalities and functional requirements for rack computing, memory-sharing computing, and resource disaggregated computing models, all of which can be extended through the application of memory-centric computing.
°ü·Ã IPR È®¾à¼­ Á¢¼öµÈ IPR È®¾à¼­ ¾øÀ½
°ü·ÃÆÄÀÏ    TTAK.KO-10.1630.pdf TTAK.KO-10.1630.pdf
Ç¥ÁØÀÌ·Â
Ç¥Áظí Ç¥ÁعøÈ£ Á¦°³Á¤ÀÏ ±¸ºÐ À¯È¿
¿©ºÎ
IPR
È®¾à¼­
ÆÄÀÏ
¸Þ¸ð¸® Á᫐ ÄÄÇ»ÆÃ - ±â´É ¿ä±¸»çÇ× TTAK.KO-10.1630 2025-12-05 Á¦Á¤ À¯È¿ ¾øÀ½ TTAK.KO-10.1630.pdf