Daichi Fujiki

Daichi Fujiki

Assistant Professor

Keio University

Daichi Fujiki is an Assistant Professor in Computer Science and Informatics at Keio University. His research interests include memory-centric computing for general and application-specific workloads, and domain-specific architectures. He is leading the Data-Centric Computing Laboratory, which develops compute capable memory architectures and custom acceleration frameworks for bioinformatics workloads.

He received Ph.D. in 2022 and M.S.Eng in 2017 from the University of Michigan, Ann Arbor, and B.E. in 2016 from Keio University, Japan.

View my full profile

興味・関心
  • In-memory Computing
  • Computing for Genomics and Precision Health
  • Privacy Preserved Computing
学歴

  • Ph.D. in Computer Science and Engineering, 2022

    University of Michigan, Ann Arbor

  • M.S.Eng. in Computer Science and Engineering, 2017

    University of Michigan, Ann Arbor

  • B.E. in Computer Science and Informatics, 2016

    Keio University

Research Topics

HW/SW Co-design
In-Memory Computing
Genome Sequencing Acceleration

Experience

 
 
 
 
 
Assisntant Professor
Keio University
4月 2022 – Present Tokyo, Japan
 
 
 
 
 
AI Research Intern
Facebook AI Research
9月 2020 – 12月 2020 Remote Location
Project: Performance Characterization and Hardware Acceleration of GNNs for Recommendation System
 
 
 
 
 
GPU Architecture Intern
Samsung SARC/ACL
6月 2019 – 8月 2019 San Jose
Project: Energy Efficient Mobile GPU Shader Architecture
 
 
 
 
 
Research Intern
NVIDIA
6月 2018 – 8月 2018 Austin
Project: Near Memory Data Transformation for Sparse Workloads
 
 
 
 
 
Instructor
Tokyo University of Technology
4月 2016 – 8月 2016 Tokyo
Taught Computer Literacy class

Recent Publications

All Publications

Multi-Layer In-Memory Computing
Daichi Fujiki, Alireza Khadem, Scott Mahlke, Reetuparna Das
PDF 引用 DOI
In-Memory Acceleration for General Data Parallel Applications
Daichi Fujiki
引用
A 2.46M Reads/s Seed-Extension Accelerator for Next-Generation Sequencing Using a String-Independent PE Array
Zhehong Wang, Tianjun Zhang, Daichi Fujiki, Arun Subramaniyan, Xiao Wu, Makoto Yasuda, Satoru Miyoshi, Masaru Kawaminami, Reetuparna Das, Satish Narayanasamy, David Blaauw
引用 DOI
In-/Near-Memory Computing
Daichi Fujiki, Xiaowei Wang, Arun Subramaniyan, Reetuparna Das
引用
In-/Near-Memory Computing
SeedEx: A Genome Sequencing Accelerator for Optimal Alignments in Subminimal Space
Daichi Fujiki, Shunhao Wu, Nathan Ozog, Kush Goliya, David Blaauw, Satish Narayanasamy, Reetuparna Das
引用 DOI
Duality Cache for Data Parallel Acceleration
Daichi Fujiki, Scott Mahlke, Reetuparna Das
引用 DOI URL
Near-Memory Data Transformation for Efficient Sparse Matrix Multi-Vector Multiplication
Daichi Fujiki, Niladrish Chatterjee, Donghyuk Lee, Mike O'Connor
引用 DOI URL
AxNoC: Low-power Approximate Network-on-Chips using Critical-Path Isolation
Various parallel applications, such as numerical convergent computation and multimedia processing, have intrinsic tolerance to …
Akram B. Ahmed, Daichi Fujiki, Hiroki Matsutani, Michihiro Koibuchi, Hideharu Amano
引用 DOI
GenAx: A Genome Sequencing Accelerator
Genomics can transform health-care through precision medicine. Plummeting sequencing costs would soon make genome testing affordable to …
Daichi Fujiki, Arun Subramaniyan, Tianjun Zhang, Yu Zeng, Reetuparna Das, David Blaauw, Satish Narayanasamy
引用 DOI
In-Memory Data Parallel Processor
Daichi Fujiki, Scott Mahlke, Reetuparna Das
引用 DOI URL
See all

Projects

4月 4, 2022
In-Memory Computing
In-Memory Computing

In-Memory Computingは、メモリの物理特性を利用し、密結合のメモリアレイ内で直接計算を行う技術です。SRAMやDRAM、次世代の不揮発性メモリなど、様々なメモリ上でのインメモリ計算技術を応用し、「データ移動コストの最小化」と「超並列計算」の双方から、データインテンシブなアプリケーションの持続的な発展を支えます。

4月 3, 2022
Systems for Genome Sequencing
Systems for Genome Sequencing

次世代シーケンサにより大量の生ゲノムデータが得られる中、その効率的な処理に焦点をあてて研究を行っています。ハイスループットシーケンサから得られるショートリードに加え、近年発展が進むNanoporeなどのロングリードシーケンサを対象に、計算強度の高いアライメントやアセンブリといった二次処理から、ベースコーラーやウイルス検出に用いられる機械学習のフレームワークまで、多面的な研究をしてます。

4月 2, 2022
Privacy Preserved Database System
Privacy Preserved Database System

最新の暗号化技術とデータベース技術を組み合わせ、センシティブなデータをクラウド上でスケーラブルに運用させ、どんなときにでも安心して使えるデータベース基盤の構築を目指し、研究を行っています。