DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kim, Hyunsung | ko |
dc.contributor.author | Yoo, Sungyeob | ko |
dc.contributor.author | Bae, Jaewan | ko |
dc.contributor.author | Bong, Kyeongryeol | ko |
dc.contributor.author | Boo, Yoonho | ko |
dc.contributor.author | Charfi, Karim | ko |
dc.contributor.author | Kim, Hyo-Eun | ko |
dc.contributor.author | Kim, Hyun Suk | ko |
dc.contributor.author | Kim, Jinseok | ko |
dc.contributor.author | Lee, Byungjae | ko |
dc.contributor.author | Lee, Jaehwan | ko |
dc.contributor.author | Shim, Myeongbo | ko |
dc.contributor.author | Shin, Sungho | ko |
dc.contributor.author | Woo, Jeong Seok | ko |
dc.contributor.author | Kim, Joo-Young | ko |
dc.contributor.author | Park, Sunghyun | ko |
dc.contributor.author | Oh, Jinwook | ko |
dc.date.accessioned | 2022-11-29T02:00:43Z | - |
dc.date.available | 2022-11-29T02:00:43Z | - |
dc.date.created | 2022-11-27 | - |
dc.date.issued | 2022-08-22 | - |
dc.identifier.citation | 2022 IEEE Hot Chips 34 Symposium, HCS 2022 | - |
dc.identifier.uri | http://hdl.handle.net/10203/301195 | - |
dc.description.abstract | We present the world's first AI-enabled high-frequency trading (HFT) system, LightTrader , which integrates the custom AI accelerators and the FPGA-based conventional HFT pipeline for the low-latency-high-throughput trading solutions with a reduced query miss rate. For better utilization, adaptive job scheduling methods are also proposed to further improve the performance, where layer-wise workload scaling and dynamic voltage-frequency scaling (DVFS) techniques progressively adjust the workloads of AI accelerators, in conjunction with the architecture support. LightTrader integrating TSMC 7nm tape-out accelerators solely achieves 6x speed-up of DNN processing and 30-50x reduction of query miss rate without the scheduling method while the scheduling scheme further improves the energy efficiency by 25% and reduces the query miss rate by 2.4x. | - |
dc.language | English | - |
dc.publisher | Institute of Electrical and Electronics Engineers Inc. | - |
dc.title | LightTrader : World's first AI-enabled High-Frequency Trading Solution with 16 TFLOPS / 64 TOPS Deep Learning Inference Accelerators | - |
dc.type | Conference | - |
dc.identifier.scopusid | 2-s2.0-85140975476 | - |
dc.type.rims | CONF | - |
dc.citation.publicationname | 2022 IEEE Hot Chips 34 Symposium, HCS 2022 | - |
dc.identifier.conferencecountry | US | - |
dc.identifier.conferencelocation | Cupertino | - |
dc.identifier.doi | 10.1109/HCS55958.2022.9895619 | - |
dc.contributor.localauthor | Kim, Joo-Young | - |
dc.contributor.nonIdAuthor | Kim, Hyunsung | - |
dc.contributor.nonIdAuthor | Yoo, Sungyeob | - |
dc.contributor.nonIdAuthor | Bae, Jaewan | - |
dc.contributor.nonIdAuthor | Bong, Kyeongryeol | - |
dc.contributor.nonIdAuthor | Boo, Yoonho | - |
dc.contributor.nonIdAuthor | Charfi, Karim | - |
dc.contributor.nonIdAuthor | Kim, Hyo-Eun | - |
dc.contributor.nonIdAuthor | Kim, Hyun Suk | - |
dc.contributor.nonIdAuthor | Kim, Jinseok | - |
dc.contributor.nonIdAuthor | Lee, Byungjae | - |
dc.contributor.nonIdAuthor | Lee, Jaehwan | - |
dc.contributor.nonIdAuthor | Shim, Myeongbo | - |
dc.contributor.nonIdAuthor | Shin, Sungho | - |
dc.contributor.nonIdAuthor | Woo, Jeong Seok | - |
dc.contributor.nonIdAuthor | Park, Sunghyun | - |
dc.contributor.nonIdAuthor | Oh, Jinwook | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.