High Performance Data Processing System for Ad-hoc Data

• Project/Area Number: 16H01715
• Research Category: Grant-in-Aid for Scientific Research (A)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Software
• Research Institution: The University of Tokyo
• Principal Investigator: Taura Kenjiro 東京大学, 大学院情報理工学系研究科, 教授 (90282714)
• Project Period (FY): 2016-04-01 – 2021-03-31
• Project Status: Completed (Fiscal Year 2021)
• Budget Amount: ¥38,870,000 (Direct Cost: ¥29,900,000、Indirect Cost: ¥8,970,000); Fiscal Year 2020: ¥9,360,000 (Direct Cost: ¥7,200,000、Indirect Cost: ¥2,160,000); Fiscal Year 2019: ¥8,970,000 (Direct Cost: ¥6,900,000、Indirect Cost: ¥2,070,000); Fiscal Year 2018: ¥7,020,000 (Direct Cost: ¥5,400,000、Indirect Cost: ¥1,620,000); Fiscal Year 2017: ¥7,410,000 (Direct Cost: ¥5,700,000、Indirect Cost: ¥1,710,000); Fiscal Year 2016: ¥6,110,000 (Direct Cost: ¥4,700,000、Indirect Cost: ¥1,410,000)
• Keywords: 大規模データ処理 / ad-hocデータ処理 / 字句解析 / 構文解析 / データ抽出 / ad‐hocデータ処理 / 演算子優先度文法 / 構文解析器生成系 / 字句解析器生成系 / ad‐hocデータ / 構文解析器生成器 / 並列構文解析 / スキャナレス構文解析 / LL(*)文法 / 構文解析生成系 / ad-hocデータ / LL(*) / 正規表現 / 高性能計算 / 大規模データ / ストリーム処理 / 解析表現文法 / ニューラル機械翻訳 / 再帰型ニューラルネットワーク / 自然言語処理 / 文字列処理 / SIMD命令

Outline of Final Research Achievements

Toward the goal of high performance text processing using parallelization and vectorization, we studied lexer (or parser) generators that generate parallelized/vectorized lexers (or parsers) from regular expressions or context free grammars. We investigate an approach that vectorizes scannerless parser and an approach that parallelizes both lexers and locally parsable (thus relatively simple-to-parallelize) parsers.

Academic Significance and Societal Importance of the Research Achievements

データ活用はSociety 5.0の要諦である. 多くの利用可能なデータはテキスト形式で保存されている(XML, JSONなど標準的な形式のものもあれば, 決まった形式のないものもある). 文字列に対するデータ処理の一番はじめの段階が字句解析または構文解析と呼ばれる, 一種のパターンマッチング処理である. 本研究はそれらを容易に, かつ高速に処理することを目指したもので, 社会で利用可能なビッグデータの増大に対して有用な貢献を果たしうる研究である.

Research Products

• [Journal Article] Plex: Scaling Parallel Lexing with Backtrack-Free Prescanning (2021)
  • Author(s): Le Li, Shigeyuki Sato, Qiheng Liu, Kenjiro Taura
  • Journal Title: 2021 IEEE International Parallel and Distributed Processing Symposium (IPDPS) Volume: 会議予稿集のためなし Pages: 693-702
  • DOI: 10.1109/ipdps49936.2021.00079
  • Peer Reviewed
• [Journal Article] LL(*) 文法に基づくスキャナレス構文解析器の提案 (2018)
  • Author(s): 井原 央翔, 佐藤 重幸, 田浦 健次朗
  • Journal Title: Cross-disciplinary workshop on computing Systems, Infrastructures, and programming (xSIG) 2018 Volume: -
  • Peer Reviewed
• [Journal Article] OPG を利用したアドホックな並列データ処理系 (2018)
  • Author(s): リュウ ケイコウ, 井原 央翔, 田浦 健次朗
  • Journal Title: 情報処理学会論文誌プログラミング（PRO） Volume: 12 Pages: 1-8
  • NAID: 170000150059
• [Journal Article] LL(*) 文法に基づくスキャナレス構文解析器の提案 (2018)
  • Author(s): 井原 央翔, 佐藤 重幸, 田浦 健次朗
  • Journal Title: xSIG 2018 workshop Volume: -
  • Peer Reviewed
• [Journal Article] Parallelized software offloading of low-level communication with user-level threads (2018)
  • Author(s): Wataru Endo and Kenjiro Taura
  • Journal Title: Proceedings of the International Conference on High Performance Computing in Asia-Pacific Region Volume: - Pages: 289-298
  • DOI: 10.1145/3149457.3149475
  • Peer Reviewed
• [Journal Article] 低レイテンシ SSD をメモリ拡張として利用したときの性能評価 (2018)
  • Author(s): 中澤 弘樹, 田浦 健次朗
  • Journal Title: xSIG 2018 workshop Volume: -
  • Peer Reviewed
• [Journal Article] Cache friendly parallelization of neural encoder-decoder models without padding on multi-core architecture. (2017)
  • Author(s): Yuchen Qiao, Kazuma Hashimoto, Akiko Eriguchi, Haixia Wang, Dongsheng Wang, Yoshimasa Tsuruoka, and Kenjiro Taura.
  • Journal Title: The 6th International Workshop on Parallel and Distributed Computing for Large Scale Machine Learning and Big Data Analytics Volume: - Pages: 437-440
  • DOI: 10.1109/ipdpsw.2017.165
  • Peer Reviewed / Open Access / Int'l Joint Research
• [Journal Article] Neural Machine Translation with Source-Side Latent Graph Parsing (2017)
  • Author(s): Kazuma Hashimoto and Yoshimasa Tsuruoka
  • Journal Title: Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing (EMNLP 2017) Volume: - Pages: 125-135
  • DOI: 10.18653/v1/d17-1012
  • Peer Reviewed
• [Journal Article] Learning to Parse and Translate Improves Neural Machine Translation (2017)
  • Author(s): Akiko Eriguchi, Yoshimasa Tsuruoka, and Kyunghyun Cho
  • Journal Title: Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (ACL2017) Volume: - Pages: 72-78
  • DOI: 10.18653/v1/p17-2012
  • Peer Reviewed / Int'l Joint Research
• [Journal Article] Cache Friendly Parallel Encoder-Decoder Model without Padding on Mulit-core Architecture (2017)
  • Author(s): Yuchen Qiao, Kenjiro Taura, Kazuma Hashimoto, Yoshimasa Tsuruoka and Akkiko Eriguchi
  • Journal Title: Proceedings of The 6th International Workshop on Parallel and Distributed Computing for Large Scale Machine Learning and Big Data Analytics Volume: -
  • Peer Reviewed / Int'l Joint Research / Acknowledgement Compliant
• [Journal Article] Low latency and resource-aware program composition for large-scale data analysis (2016)
  • Author(s): Masahiro Tanaka, Kenjiro Taura, and Kentaro Torisawa
  • Journal Title: 16th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGrid) Volume: - Pages: 325-330
  • DOI: 10.1109/ccgrid.2016.88
  • Peer Reviewed
• [Journal Article] A static cut-off for task parallel programs (2016)
  • Author(s): Shintaro Iwasaki, Kenjiro Taura
  • Journal Title: Proceedings of the 2016 International Conference on Parallel Architectures and Compilation Volume: - Pages: 139-150
  • DOI: 10.1145/2967938.2967968
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Fragmented BWT: An Extended BWT for Full-Text Indexing (2016)
  • Author(s): Masaru Ito, Hiroshi Inoue, and Kenjiro Taura
  • Journal Title: International Symposium on String Processing and Information Retrieval Volume: - Pages: 97-109
  • DOI: 10.1007/978-3-319-46049-9_10
  • ISBN: 9783319460482, 9783319460499
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Autotuning of a Cut-Off for Task Parallel Programs (2016)
  • Author(s): Shintaro Iwasaki, Kenjiro Taura
  • Journal Title: IEEE 10th International Symposium on Embedded Multicore/Many-core Systems-on-Chip (MCSoC) Volume: - Pages: 353-360
  • DOI: 10.1109/mcsoc.2016.51
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Domain Adaptation and Attention-Based Unknown Word Replacement in Chinese-to-Japanese Neural Machine Translation (2016)
  • Author(s): Kazuma Hashimoto, Akiko Eriguchi, and Yoshimasa Tsuruoka
  • Journal Title: the 3rd Workshop on Asian Translation (WAT2016) Volume: - Pages: 75-83
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Character-based Decoding in Tree-to-Sequence Attention-based Neural Machine Translation (2016)
  • Author(s): Akiko Eriguchi, Kazuma Hashimoto, and Yoshimasa Tsuruoka
  • Journal Title: the 3rd Workshop on Asian Translation (WAT2016) Volume: - Pages: 175-183
  • Peer Reviewed / Acknowledgement Compliant
• [Presentation] Simultaneous Finite Automaton の部分構成による並列正規表現マッチ (2022)
  • Author(s): 高品 剛大, 佐藤 重幸, 田浦 健次朗
  • Organizer: 第24回プログラミングおよびプログラミング言語ワークショップ（PPL 2022）
• [Presentation] OPGを使ったアドホックな大規模な文字列データ解析のための並列処理系 (2019)
  • Author(s): リュウ ケイコウ
  • Organizer: 情報処理学会論文誌プログラミング（PRO）
• [Presentation] A Quest for Unified, Global View Parallel Programming Models for Our Future (2016)
  • Author(s): Kenjiro Taura
  • Organizer: A Quest for Unified, Global View Parallel Programming Models for Our Future
  • Place of Presentation: Kyoto
  • Int'l Joint Research / Invited