Realizing Fast and Scalable Brain Information Processing based on Skeletal Parallel Programming

• Project/Area Number: 25330088
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Software
• Research Institution: Kochi University of Technology
• Principal Investigator: Matsuzaki Kiminori 高知工科大学, 工学部, 准教授 (30401243)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Project Status: Completed (Fiscal Year 2015)
• Budget Amount: ¥4,680,000 (Direct Cost: ¥3,600,000、Indirect Cost: ¥1,080,000); Fiscal Year 2015: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000); Fiscal Year 2014: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000); Fiscal Year 2013: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
• Keywords: 並列プログラミング / 脳情報処理 / 計算パターン / 超解像処理 / MapReduce / ソフトウェア / fMRI / 超解像 / 並列化 / 脳科学

Outline of Final Research Achievements

This research is to apply systematic parallel programming methodology to applications in the brain science area. It consists of the following two tasks: task 1) application of parallel programming to brain image processing applications; task 2) study on large-scale parallel/distributed framework for brain-image database. We applied three program-optimization methods to super-resolution processing of fMRI images and obtained speedups by a factor of about three by each of methods. We also proposed an algorithm to process data with tree dependency on the MapReduce framework and formalized the functional model of Hadoop MapReduce.

Research Products

• [Journal Article] Functional Models of Hadoop MapReduce with Application to Scan (2016)
  • Author(s): Kiminori Matsuzaki
  • Journal Title: International Journal of Parallel Programming Volume: (online) Issue: 2 Pages: 362-381
  • DOI: 10.1007/s10766-016-0414-9
  • NAID: 110009850860
  • Peer Reviewed
• [Journal Article] fMRI画像に対する超解像処理とその効果についての検証 (2015)
  • Author(s): 松崎 公紀，宮崎 玲奈
  • Journal Title: 高知工科大学紀要 Volume: 12
  • Open Access / Acknowledgement Compliant
• [Journal Article] An Automatic Fusion Mechanism for Variable-Length List Skeletons in SkeTo (2014)
  • Author(s): Kento Emoto, Kiminori Matsuzaki
  • Journal Title: International Journal of Parallel Programming Volume: 42(4) Issue: 4 Pages: 546-563
  • DOI: 10.1007/s10766-013-0263-8
  • NAID: 120005728994
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Accumulative Computation on MapReduce (2014)
  • Author(s): Yu Liu, Kento Emoto, Kiminori Matsuzaki, Zhenjiang Hu
  • Journal Title: 情報処理学会論文誌プログラミング Volume: 7 Pages: 18-27
  • NAID: 130003394611
  • Peer Reviewed
• [Presentation] プログラム高速化の効果の検証 －超解像処理を題材として (2016)
  • Author(s): 畑中 涼, 松崎 公紀
  • Organizer: 第18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（PPL2016）
  • Place of Presentation: 岡山県
  • Year and Date: 2016-03-07
• [Presentation] MR画像に対する超解像化とその高速化に関する検討 (2016)
  • Author(s): 松崎 公紀
  • Organizer: 医用画像情報学会平成27年度春季(第174回)大会
  • Place of Presentation: 広島県
  • Year and Date: 2016-02-06
• [Presentation] MRI Image Processing with OpenCL (2015)
  • Author(s): Juan Luis Soler Ferrer, Kiminori Matsuzaki
  • Organizer: 第32回日本ソフトウェア科学会大会
  • Place of Presentation: 東京都
  • Year and Date: 2015-09-08
• [Presentation] Parallel Tree Accumulations on MapReduce (2014)
  • Author(s): Kiminori Matsuzaki, Reina Miyazaki
  • Organizer: 7th International Symposium on High-level Parallel Programming and Applications (HLPP 2014)
  • Place of Presentation: アムステルダム，オランダ
  • Year and Date: 2014-07-03 – 2014-07-04
• [Presentation] A Generic Implementation of Tree Skeletons (2014)
  • Author(s): Shigeyuki Sato, Kiminori Matsuzaki
  • Organizer: 7th International Symposium on High-level Parallel Programming and Applications (HLPP 2014)
  • Place of Presentation: アムステルダム，オランダ
  • Year and Date: 2014-07-03 – 2014-07-04
• [Presentation] A Tree-Division Tool for Practical Huge XML Trees (Poster) (2013)
  • Author(s): Kawamura Takayuki, Kiminori Matsuzaki
  • Organizer: 11th Asian Symposium on Programming Languages and Systems (APLAS2013)
  • Place of Presentation: メルボルン，オーストラリア
• [Presentation] An Automatic Fusion Mechanism for Variable-Length List Skeletons in SkeTo (2013)
  • Author(s): Kento Emoto, Kiminori Matsuzaki
  • Organizer: International Symposium on High-level Parallel Programming and Applications
  • Place of Presentation: パリ，フランス