https://scholars.lib.ntu.edu.tw/handle/123456789/632317
標題: | RM-SSD: In-Storage Computing for Large-Scale Recommendation Inference | 作者: | Sun X Wan H Li Q CHIA-LIN YANG TEI-WEI KUO Xue C.J. |
關鍵字: | n/a | 公開日期: | 2022 | 卷: | 2022-April | 起(迄)頁: | 1056-1070 | 來源出版物: | Proceedings - International Symposium on High-Performance Computer Architecture | 摘要: | To meet the strict service level agreement requirements of recommendation systems, the entire set of embeddings in recommendation systems needs to be loaded into the memory. However, as the model and dataset for production-scale recommendation systems scale up, the size of the embeddings is approaching the limit of memory capacity. Limited physical memory constrains the algorithms that can be trained and deployed, posing a severe challenge for deploying advanced recommendation systems. Recent studies offload the embedding lookups into SSDs, which targets the embedding-dominated recommendation models. This paper takes it one step further and proposes to offload the entire recommendation system into SSD with in-storage computing capability. The proposed SSD-side FPGA solution leverages a low-end FPGA to speed up both the embedding-dominated and MLP-dominated models with high resource efficiency. We evaluate the performance of the proposed solution with a prototype SSD. Results show that we can achieve 20-100× throughput improvement compared with the baseline SSD and 1.5-15× improvement compared with the state-of-art. © 2022 IEEE. |
URI: | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85130735559&doi=10.1109%2fHPCA53966.2022.00081&partnerID=40&md5=c99f9d855c9187a4bb520ab4b5598d3a https://scholars.lib.ntu.edu.tw/handle/123456789/632317 |
ISSN: | 15300897 | DOI: | 10.1109/HPCA53966.2022.00081 | SDG/關鍵字: | Embeddings; Field programmable gate arrays (FPGA); Computing capability; Embeddings; Large-scales; Lookups; Memory capacity; N/a; Physical memory; Production scale; Scale-up; Servicelevel agreement (SLA); Recommender systems |
顯示於: | 資訊工程學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。