{"id":139876,"date":"2023-05-25T17:23:24","date_gmt":"2023-05-25T08:23:24","guid":{"rendered":"http:\/\/ee.presscat.kr\/?post_type=research-achieve&p=139876"},"modified":"2026-04-13T16:33:16","modified_gmt":"2026-04-13T07:33:16","slug":"ee-prof-myoungsoo-jungs-research-team-develops-the-worlds-first-ai-semiconductor-for-search-engines-based-on-cxl-3-0","status":"publish","type":"research-achieve","link":"http:\/\/ee.presscat.kr\/en\/research-achieve\/ee-prof-myoungsoo-jungs-research-team-develops-the-worlds-first-ai-semiconductor-for-search-engines-based-on-cxl-3-0\/","title":{"rendered":"EE Prof. Myoungsoo Jung\u2019s research team develops the world’s first AI semiconductor for search engines based on CXL 3.0."},"content":{"rendered":"

Our department\u2019s Professor Myounsoo Jung\u2019s research team has developed the world\u2019s first AI semiconductor for search engines based on CXL 3.0.<\/span><\/p>\n

 <\/p>\n

Approximate nearest neighbor search (ANNS) is widely used in commercial services such as image search, database, and recommendation systems. <\/span><\/p>\n

However, in production-level ANNS, there is a challenge of requiring a large amount of memory due to the extensive dataset. <\/span><\/p>\n

To address this memory pressure issue, modern ANNS techniques leverage lossy compression methods or employ persistent storage for their memory expansion. <\/span><\/p>\n

However, these approaches often suffer from low accuracy and performance.<\/span><\/p>\n

 <\/p>\n

The research team proposed expanding memory capacity via compute express link (CXL), which is PCIe based open-industry interconnect technology that allows the underlying working memory to be highly scalable and composable at a low cost. <\/span><\/p>\n

Furthermore, the use of a CXL switch enables connecting multiple memory expanders to a single port, providing greater scalability. However, memory expansion through CXL has the disadvantage of increased memory access time compared to local memory.<\/span><\/p>\n

 <\/p>\n

The research team has developed an AI semiconductor, ‘CXL-ANNS<\/b>‘, which leverages CXL switch and memory expanders to accommodate high memory pressure that comes from extensive datasets without losing accuracy or performance. <\/span><\/p>\n

Additionally, by using near data processing and data partitioning based on locality, the performance of CXL-ANNS is improved. <\/span><\/p>\n

They also compared prototyped CXL-ANNS with the existing solutions for ANNS. Compared to previous research, CXL-ANNS shows 111 times<\/b> higher performance. Particularly, 92 times<\/b> higher performance can be achieved compared to Microsoft’s solution that is used in commercial service.<\/span><\/p>\n

 <\/p>\n

This research, along with the paper titled “CXL-ANNS: Software-Hardware Collaborative Memory Disaggregation and Computation for Billion-Scale Approximate Nearest Neighbor Search”, will be presented in July at \u2018USENIX Annual Technical Conference, ATC, 2023\u2019.<\/span><\/p>\n

 <\/p>\n

\"\"<\/span><\/p>\n

 <\/p>\n

\"\"<\/span><\/p>\n

 <\/p>\n

\"\"<\/span><\/p>\n

 <\/p>\n

The research was supported by Panmnesia (http:\/\/panmnesia.com)<\/u><\/a>. More information on this paper can be found at CAMELab website (http:\/\/camelab.org).<\/span><\/p>\n

 <\/p>\n

[News Link]<\/span><\/p>\n

The Korea Economic Daily: https:\/\/www.hankyung.com\/it\/article\/202305259204i<\/span><\/p>\n

The Herald Business: http:\/\/news.heraldcorp.com\/view.php?ud=20230525000225<\/span><\/p>\n

ChosunBiz: https:\/\/biz.chosun.com\/science-chosun\/technology\/2023\/05\/25\/4UW5LPX3WVARVIS3QBBICPINFM\/<\/span><\/p>\n

etnews: https:\/\/www.etnews.com\/20230525000092<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"

398<\/p>\n","protected":false},"featured_media":139874,"template":"","research_category":[],"class_list":["post-139876","research-achieve","type-research-achieve","status-publish","has-post-thumbnail","hentry"],"acf":[],"_links":{"self":[{"href":"http:\/\/ee.presscat.kr\/en\/wp-json\/wp\/v2\/research-achieve\/139876","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/ee.presscat.kr\/en\/wp-json\/wp\/v2\/research-achieve"}],"about":[{"href":"http:\/\/ee.presscat.kr\/en\/wp-json\/wp\/v2\/types\/research-achieve"}],"wp:featuredmedia":[{"embeddable":true,"href":"http:\/\/ee.presscat.kr\/en\/wp-json\/wp\/v2\/media\/139874"}],"wp:attachment":[{"href":"http:\/\/ee.presscat.kr\/en\/wp-json\/wp\/v2\/media?parent=139876"}],"wp:term":[{"taxonomy":"research_category","embeddable":true,"href":"http:\/\/ee.presscat.kr\/en\/wp-json\/wp\/v2\/research_category?post=139876"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}