LightTrader : World's first AI-enabled High-Frequency Trading Solution with 16 TFLOPS / 64 TOPS Deep Learning Inference Accelerators

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 706
  • Download : 0
DC FieldValueLanguage
dc.contributor.authorKim, Hyunsungko
dc.contributor.authorYoo, Sungyeobko
dc.contributor.authorBae, Jaewanko
dc.contributor.authorBong, Kyeongryeolko
dc.contributor.authorBoo, Yoonhoko
dc.contributor.authorCharfi, Karimko
dc.contributor.authorKim, Hyo-Eunko
dc.contributor.authorKim, Hyun Sukko
dc.contributor.authorKim, Jinseokko
dc.contributor.authorLee, Byungjaeko
dc.contributor.authorLee, Jaehwanko
dc.contributor.authorShim, Myeongboko
dc.contributor.authorShin, Sunghoko
dc.contributor.authorWoo, Jeong Seokko
dc.contributor.authorKim, Joo-Youngko
dc.contributor.authorPark, Sunghyunko
dc.contributor.authorOh, Jinwookko
dc.date.accessioned2022-11-29T02:00:43Z-
dc.date.available2022-11-29T02:00:43Z-
dc.date.created2022-11-27-
dc.date.issued2022-08-22-
dc.identifier.citation2022 IEEE Hot Chips 34 Symposium, HCS 2022-
dc.identifier.urihttp://hdl.handle.net/10203/301195-
dc.description.abstractWe 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.languageEnglish-
dc.publisherInstitute of Electrical and Electronics Engineers Inc.-
dc.titleLightTrader : World's first AI-enabled High-Frequency Trading Solution with 16 TFLOPS / 64 TOPS Deep Learning Inference Accelerators-
dc.typeConference-
dc.identifier.scopusid2-s2.0-85140975476-
dc.type.rimsCONF-
dc.citation.publicationname2022 IEEE Hot Chips 34 Symposium, HCS 2022-
dc.identifier.conferencecountryUS-
dc.identifier.conferencelocationCupertino-
dc.identifier.doi10.1109/HCS55958.2022.9895619-
dc.contributor.localauthorKim, Joo-Young-
dc.contributor.nonIdAuthorKim, Hyunsung-
dc.contributor.nonIdAuthorYoo, Sungyeob-
dc.contributor.nonIdAuthorBae, Jaewan-
dc.contributor.nonIdAuthorBong, Kyeongryeol-
dc.contributor.nonIdAuthorBoo, Yoonho-
dc.contributor.nonIdAuthorCharfi, Karim-
dc.contributor.nonIdAuthorKim, Hyo-Eun-
dc.contributor.nonIdAuthorKim, Hyun Suk-
dc.contributor.nonIdAuthorKim, Jinseok-
dc.contributor.nonIdAuthorLee, Byungjae-
dc.contributor.nonIdAuthorLee, Jaehwan-
dc.contributor.nonIdAuthorShim, Myeongbo-
dc.contributor.nonIdAuthorShin, Sungho-
dc.contributor.nonIdAuthorWoo, Jeong Seok-
dc.contributor.nonIdAuthorPark, Sunghyun-
dc.contributor.nonIdAuthorOh, Jinwook-
Appears in Collection
EE-Conference Papers(학술회의논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0