Kazuaki Ishizaki
(石崎　一明)           

• RaPiD: AI Accelerator for Ultra-low Precision Training and Inference [IEEE Xplore]
  Swagath Venkataramani, Vijayalakshmi Srinivasan, Wei Wang, Sanchari Sen, Jintao Zhang, Ankur Agrawal, Monodeep Kar, Shubham Jain, Alberto Mannari, Hoang Tran, Yulong Li, Eri Ogawa, Kazuaki Ishizaki, Hiroshi Inoue, Marcel Schaal, Mauricio Serrano, Jungwook Choi, Xiao Sun, Naigang Wang, Chia-Yu Chen, Allison Allain, James Bonano, Nianzheng Cao, Robert Casatuta, Matthew Cohen, Bruce Fleischer, Michael Guillorn, Howard Haynie, Jinwook Jung, Mingu Kang, Kyu-hyoun Kim, Siyu Koswatta, Saekyu Lee, Martin Lutz, Silvia Mueller, Jinwook Oh, Ashish Ranjan, Zhibin Ren, Scot Rider, Kerstin Schelm, Michael Scheuermann, Joel Silberman, Jie Yang, Vidhi Zalani, Xin Zhang, Ching Zhou, Matt Ziegler, Vinay Shah, Moriyoshi Ohara, Pong-Fei Lu, Brian Curran, Sunil Shukla, Leland Chang, and Kailash Gopalakrishnan
  The International Symposium on Computer Architecture (ISCA),  2021
• A Compiler for Deep Neural Network Accelerators to Generate Optimized Code for a Wide Range of Data Parameters from a Hand-crafted Computation Kernel [IEEE Xplore]
  Eri Ogawa, Kazuaki Ishizaki, Hiroshi Inoue, Swagath Venkataramani, Jungwook Choi, Wei Wang, Vijayalakshmi Srinivasan, Moriyoshi Ohara and Kailash Gopalakrishnan
  IEEE Symposium on Low-Power and High-Speed Chips and Systems (COOL Chips 22),  2019
• Analyzing and Optimizing Java Code Generation for Apache Spark Query Plan [ACM Author-Izer download][slide]
  Kazuaki Ishizaki
  The the 10th ACM/SPEC International Conference on Performance Engineering (ICPE2019),  2019
• Identifying the potential of near data processing for apache spark [ACM Author-Izer download]
  Ahsan Javed Awan, Moriyoshi Ohara, Eduard Ayguadé, Kazuaki Ishizaki, Mats Brorsson, Vladimir Vlassov:
  The 4th Annual International Symposium on Memory Systems (MEMSYS 2017), 2017
• Accelerating Spark Datasets by inlining deserialization [link][slide]
  Jan Wróblewski, Kazuaki Ishizaki, Hiroshi Inoue, and Moriyoshi Ohara
  The 31st IEEE International Parallel & Distributed Processing Symposium (IPDPS 2017),  2017
• Compiling and Optimizing Java 8 Programs for GPU Execution [pdf][slide]
  Kazuaki Ishizaki, Akihiro Hayashi, Gita Koblents, and Vivek Sarkar
  The 24th International Conference on Parallel Architectures and Compilation Techniques (PACT-2015),  2015
• Machine-Learning-based Performance Heuristics for Runtime CPU/GPU Selection.
  [ACM Author-Izer download] [slide]
  Akihiro Hayashi, Kazuaki Ishizaki, Gita Koblents, Vivek Sarkar
  12th International Conference on the Principles and Practice of Programming in Java (PPPJ), 2015
• On the Benefits and Pitfalls of Extending a Statically Typed Language JIT Compiler for Dynamic Scripting Languages [ACM Author-Izer download]
  Jose Castanos, David Edelsohn, Kazuaki Ishizaki, Priya Nagpurkar, Toshio Nakatani, Takeshi Ogasawara, and Peng Wu
  2012 ACM Object-Oriented Programming, Systems, Languages & Applications (SPLASH/OOPSLA 2012), 2012
• Adding dynamically-typed language support to a statically-typed language compiler: performance evaluation, analysis, and tradeoffs [ACM Author-Izer download]
  Kazuaki Ishizaki, Takeshi Ogasawara, Jose Castanos, Priya Nagpurkar, David Edelsohn, and Toshio Nakatani
  Eighth Annual International Conference on Virtual Execution Environments (VEE 2012), 2012
• Distributed and Fault-Tolerant Execution Framework for Transaction Processing [ACM Author-Izer download]
  Toshio Suganuma, Akira Koseki, Kazuaki Ishizaki, Yohei Ueda, Ken Mizuno, Daniel Silva, Hideaki Komatsu, and Toshio Nakatani
  The 4th Annual Haifa Experimental Systems Conference (SYSTOR 2011), 2011

• Parallel Programming Framework for Large Batch Transaction Processing on Scale-out Systems [ACM Author-Izer download]
  Kazuaki Ishizaki, Toshio Suganuma, Akira Koseki, Yohei Ueda, Ken Mizuno, Daniel Silva, Hideaki Komatsu, Toshio Nakatani
  The 3rd Annual Haifa Experimental Systems Conference (SYSTOR 2010), Article No. 15, 2010
• Analyzing and Improving Performance Scalability of Commercial Server Workloads on a Chip Multiprocessor [link]
  Kazuaki Ishizaki, Shahrokh Daijavad, Toshio Nakatani
  2009 IEEE International Symposium on Workload Characterization (IISWC 2009), pp. 217-226, 2009
• The Data-centricity of Web 2.0 Workloads and its Impact on Server Performance [link]
  Moriyoshi Ohara, Priya Nagpurkar, Yohei Ueda, Kazuaki Ishizaki
  2009 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS 2009), pp. 133-142, 2009
• Effectiveness of Cross-Platform Optimizations for a Java Just-In-Time Compiler [pdf] (396KB)[ACM Author-Izer download]
  Kazuaki Ishizaki, Mikio Takeuchi, Kiyokuni Kawachiya, Toshio Suganuma, Osamu Gohda, Tatsushi Inagaki, Akira Koseki, Kazunori Ogata, Motohiro Kawahito, Toshiaki Yasue, Takeshi Ogasawara, Tamiya Onodera, Hideaki Komatsu, and Toshio Nakatani
  ACM Conference on Object-Oriented Programming, Systems, Languages, and Applications (OOPSLA 2003), pp.187-204, 2003
• Eliminating Exception Constraints of Java Programs for IA-64 [pdf] (252KB) [link]
  Kazuaki Ishizaki, Tatsushi Inagaki, Hideaki Komatsu, and Toshio Nakatani
  The Eleventh International Conference on Parallel Architectures and Compilation Techniques (PACT-2002), pp. 259-268, 2002
• A Study of Devirtualization Techniques for a Java Just-In-Time Compiler [pdf] (221KB) [ACM Author-Izer download]
  Kazuaki Ishizaki, Motohiro Kawahito, Toshiaki Yasue, Hideaki Komatsu, and Toshio Nakatani,
  ACM SIGPLAN Conference on Object-Oriented Programming Systems, Languages, and Applications (OOPSLA'00), 2000
• Design, Implementation, and Evaluation of Optimizations in a Just-In-Time Compiler [pdf] (1MB) [ACM Author-Izer download]
  Kazuaki Ishizaki, Motohiro Kawahito, Toshiaki Yasue, Mikio Takeuchi, Takeshi Ogasawara, Toshio Suganuma, Tamiya Onodera, Hideaki Komatsu, and Toshio Nakatani
  1999 ACM JavaGrande Conference, pp.119-128, 1999
• An Algorithm for Automatic Detection of Loop Indices for Communication Overlapping [link]
  International Symposium on High Performance Computing
  Kazuaki Ishizaki, Hideaki Komatsu, and Toshio Nakatani
  Lecture Notes In Computer Science Vol. 1336, pp.217-230, Springer-Verlag, 1997
• A Compiler Algorithm for Overlapping Communication with Computation on Distributed Memory Machines
  Kazuaki Ishizaki and Hideaki Komatsu
  Joint Symposium on Parallel Processing (JSPP1996), pp.339-346, 1996 (in Japanese)

Back to top

• Workload and network-optimized computing systems [link]
  David P. LaPotin, Shahrokh Daijavad, Charles L. Johnson, Steven W. Hunter, Kazuaki Ishizaki, Hubertus Franke, Heather D. Achilles, Dan P. Dumarot, Nancy A. Greco, and Bijan Davari
  IBM Journal of Research and Development, Vol. 54, No. 1, pp. 1:1-1:12, 2010
• A Programming Framework for Parallel Transactional Applications [link]
  Ken Mizuno, Toshio Suganuma, Kazuaki Ishizaki, Akira Koseki, Yohei Ueda, Hideaki Komatsu
  Information Processing Society of Japan Transactions on Programming, Vol. 48, No. SIG12(PRO 34), pp. 41-51, 2007 (in Japanese)
• Evolution of a Java just-in-time compiler for IA-32 platforms [pdf] (265KB) [link]
  Toshio Suganuma, Takeshi Ogasawara, Kiyokuni Kawachiya, Mikio Takeuchi, Kazuaki Ishizaki, Akira Koseki, Tatsushi Inagaki, Toshiaki Yasue, Motohiro Kawahito, Tamiya Onodera, Hideaki Komatsu, and Toshio Nakatani
  IBM Journal of Research and Development, IBM Research in Asia Issue, Vol. 48, No. 5/6, pp. 767-795, 2004
• Eliminating Exception Constraints of Java Programs for IA-64 [link]
  Kazuaki Ishizaki, Tatsushi Inagaki, Hideaki Komatsu, and Toshio Nakatani
  The Journal of Instruction Level Parallelism, Vol. 5, pp. 1-22, 2003
• Eliminating Exception Constraints in Java Programs Using a Speculation Technique [link]
  Kazuaki Ishizaki, Tatsushi Inagaki, and Hideaki Komatsu
  Information Processing Society of Japan Transactions on Programming, Vol. 43, No. SIG8(PRO 15), pp. 87-97, 2002 (in Japanese)
• A Direct Devirtualization Technique with the Code Patching Mechanism [link]
  Kazuaki Ishizaki, Toshiaki Yasue, Motohiro Kawahito, and Hideaki Komatsu
  Information Processing Society of Japan Journal, Vol. 43, No. 1, pp. 124-136, 2002 (in Japanese)
• Design, Implementation, and Evaluation of Optimizations in a Just-In-Time Compiler
  Kazuaki Ishizaki, Motohiro Kawahito, Toshiaki Yasue, Mikio Takeuchi, Takeshi Ogasawara, Toshio Suganuma, Tamiya Onodera, Hideaki Komatsu, and Toshio Nakatani
  Concurrency: Practice and Experience, Vol. 12, No. 6, pp. 457-475, 2000
• A Loop Transformation Algorithm for Communication Overlapping
  Kazuaki Ishizaki, Hideaki Komatsu, Toshio Nakatani
  International Journal of Parallel Processing, Vol. 28, No. 2, pp.135-154, 2000
• Overview of the IBM Java Just-In-Time Compiler [pdf] (966KB) [link]
  Toshio Suganuma, Takeshi Ogasawara, Mikio Takeuchi, Toshiaki Yasue, Motohiro Kawahito, Kazuaki Ishizaki, Hideaki Komatsu, and Toshio Nakatani
  IBM Systems Journals, Java Performance Issue, Vol. 39, No. 1, 2000.
• A Compiler Algorithm for Overlapping Communication with Computation on Distributed Memory Machines [link]
  Kazuaki Ishizaki and Hideaki Komatsu
  Information Processing Society of Japan Journal, Vol. 38, No. 9, pp.1849-1858, 1997 (in Japanese)

Back to top

• A Compiler Overview for A Deep Neural Network Accelerator with Software-Controlled Memory Hierarchies
  Kazuaki Ishizaki, Eri Ogawa, Hiroshi Inoue, Swagath Venkataramani, Jintao Zhang, Wei Wang, Vijayalakshmi Srinivasan, Moriyoshi Ohara, Kailash Gopalakrishnan
  IPSJ SIG PRO, 2020
• Verification of memory access for deep learning compilers [link]
  Tung D. Le, Eri Ogawa, Hiroshi Inoue, Kazuaki Ishizaki, Swagath Venkataramani, and Moriyoshi Ohara
  36th JSSST Annual Conference, 一般3-1, 2019
  
• Enhancements of Code Generation in Apache Spark [link]
  Kazuaki Ishizaki
  33rd JSSST Annual Conference, 一般3-1, 2016 (in Japanese)
  
• How Java runtime can execute practical Java programs on GPU [link]
  Kazuaki Ishizaki, Gita Koblents, Akihiro Hayashi, Vivek Sarkar, and Hiroshi Inoue
  2015 Symposium on High Performance Computing and Computational Science (HPCS 2015), pp. 96, 2015
• Transforming Java programs for concurrency using Double-Checked Locking pattern [link]
  Kazuaki Ishizaki, Shahrokh Daijavad, Toshio Nakatani
  2014 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS 2014), pp. 128-129, 2014
  
• On the Benefits and Pitfalls of Extending a Statically Typed Language JIT Compiler for Dynamic Scripting Languages
  Jose Castanos, David Edelsohn, Kazuaki Ishizaki,Toshio Nakatani, Takeshi Ogasawara, Priya Nagpurkar, Peng Wu
  IBM Programming Languages Day, 2012
• Refactoring Java Programs using Concurrent Libraries [ACM Author-Izer download
  Kazuaki Ishizaki, Shahrokh Daijavad, Toshio Nakatani
  Workshop on Parallel and Distributed Systems: Testing, Analysis, and Debugging (PADTAD - IX), 2011
• Compilers are from Mars, Dynamic Scripting Languages are from Venus
  Jose Castanos, David Edelsohn, Kazuaki Ishizaki, Priya Nagpurkar, Takeshi Ogasawara, Akihiko Tozawa, and Peng Wu
  9th Workshop on Compiler-Driven Performance, 2010
• Compilers are from Mars, Dynamic Scripting Languages are from Venus
  Jose Castanos, David Edelsohn, Kazuaki Ishizaki, Takeshi Ogasawara, Priya Nagpurkar, Akihiko Tozawa, and Peng Wu
  IBM Programming Languages Day, 2010
• Whole-Stack Analysis and Optimization of Commercial Workloads on Server Systems
  C. Richard Attanasio, Jong-Deok Choi, Niteesh Dubey, Kattamuri Ekanadham, Manish Gupta, Tatsushi Inagaki, Kazuaki Ishizaki, Joefon Jann, Robert D. Johnson, Toshio Nakatani, Il Park, Pratap Pattnaik, Mauricio J. Serrano, Stephen E. Smith, Ian Steiner, and Yefim Shuf
  Network and Parallel Computing (NPC2004), 2004
• Effectiveness of Cross-Platform Optimizations for a Java Just-In-Time Compiler [pdf]
  Kazuaki Ishizaki
  The 6th JSSST Workshop on Programming and Programming Languages, 2004 (in Japanese)
• Fast Predicate-Aware Dataflow Analysis [slide online]
  Kazuaki Ishizaki, Akira Koseki, Hideaki Komatsu, and Toshio Nakatani
  Third Workshop on Explicitly Parallel Instruction Computing Architectures and Compiler Technology (EPIC-3), 2004
• A Direct Devirtualization Technique with the Code Patching Mechanism
  Kazuaki Ishizaki, Toshiaki Yasue, Motohiro Kawahito, and Hideaki Komatsu
  IPSJ SIG PRO, 2002 (in Japanese)
• A study of Devirtualization Techniques for a Java(tm) Just-In-Time Compiler
  Kazuaki Ishizaki
  Dagstule Seminars 00451: Effective Implementation of Object-Oriented Programming Languages, 2000
• Design, Implementation and Evaluation of Optimizations in a Java Just-In-Time Compiler
  Kazuaki Ishizaki, Motohiro Kawahito, Kazuhiro Konno, Toshiaki Yasue, Mikio Takeuchi, Takeshi Ogasawara, Toshio Suganuma, Tamiya Onodera, and Hideaki Komatsu
  IEICE SIG Technical Report, CPSY-99-64, pp. 17-24, 1999 (in Japanese)
• A Loop Parallelization Algorithm for HPF Compilers
  Kazuaki Ishizaki, Hideaki Komatsu
  Eighth Annual Workshop on Language and Compilers for Parallel Computing, pp. 12.1-15, 1995
  Lecture Notes In Computer Science Vol. 1033, pp. 176-190, Springer-Verlag, 1995
• A Loop Parallelization Technique for HPF Compilers [link]
  Kazuaki Ishizaki and Hideaki Komatsu
  IPSJ SIG Technical Report, HPC-57-18, pp. 103-108, 1995 (in Japanese)
• A method for reducing the overhead of dynamic communication analysis [link]
  Takeshi Ogasawara, Kazuaki Ishizaki, and Hideaki Komatsu
  IPSJ SIG Technical Report, HPC-57-19, pp. 109-114, 1995 (in Japanese)
• Design, Implementation, and Evaluation of an HPF Compiler [link]
  Osamu Gohda, Gyoh Ohsawa, Hideaki Komatsu, Toshio Sugamuna, Takeshi Ogasawara, Kazuaki Ishizaki, and Toshio Nakatani
  IPSJ SIG Technical Report, HPC-57-20, pp. 115-120, 1995 (in Japanese)
• A loop parallelization algorithm for distributed memory machines
  Kazuaki Ishizaki and Hideaki Komatsu
  IEICE SIG Technical Report, CPSY-384, pp.97-104, 1994 (in Japanese)
• A Control Scheme of Processing Element for the Parallel Processing System -Harray-
  Yoshihiko Ishii, Kazuaki Ishizaki, Takeshi Hagimoto, Hayato Yamana, and Yoichi Muraoka
  The 43th Annual Conference of IPSJ, 1991 (in Japanese)
• A Parallel Execution Scheme of Conditional Branches Using Eager Evaluation for the Parallel Processing System -Harray- [link]
  Hayato Yamana, Kazuaki Ishizaki, Toshiaki Yasue, and Yoichi Muraoka
  IPSJ SIG Technical Report, ARC-89-19, 1991 (in Japanese)
• A Network Construction of Parallel Processor for Eager Evaluation [link]
  Kazuaki Ishizaki, Toshiaki Yasue, Hayato Yamana, and Yoichi Muraoka
  IPSJ SIG Technical Report, ARC-89-20, 1991 (in Japanese)
• A Control Scheme of Processing Element for the Parallel Processing System -Harray-
  Kazuaki Ishizaki, Yoshihiko Ishii, Takeshi Hagimoto, Hayato Yamana, and Yoichi Muraoka
  The 42th Annual Conference of IPSJ, 1991 (in Japanese)
• A Construction of Switching Unit for Parallel Processing System -Harray- [link]
  Kazuaki Ishizaki, Hayato Yamana, and Yoichi Muraoka
  IPSJ SIG Technical Report, ARC-85-2, pp. 7-14, 1990 (in Japanese)
• An Execution Scheme of a Function for Parallel Processing System -Harray-
  Kazuaki Ishizaki, Jun Kohdate, Hayato Yamana, and Yoichi Muraoka
  IEICE SIG Technical Report, CPSY-90-2, pp. 7-14, 1990 (in Japanese)

Back to top

• Optimization Techniques in a Java Just-In-Time Compiler (in Japanese) [pdf]
  Ph.D. Thesis, Waseda University, 2002
  Presentation (in Japanese) [pdf] , Abstract (in Japanese) [pdf], Abstract [pdf]

Back to top

• Efficient AI System Design With Cross-Layer Approximate Computing [link]
  Swagath Venkataramani, Xiao Sun, Naigang Wang, Chia-Yu Chen, Jungwook Choi, Mingu Kang, Ankur Agarwal, Jinwook Oh, Shubham Jain, Tina Babinsky, Nianzheng Cao, Thomas Fox, Bruce Fleischer, George Gristede, Michael Guillorn, Howard Haynie, Hiroshi Inoue, Kazuaki Ishizaki, Michael Klaiber, Shih-Hsien Lo, Gary Maier, Silvia Mueller, Michael Scheuermann, Eri Ogawa, Marcel Schaal, Mauricio Serrano, Joel Silberman, Christos Vezyrtzis, Wei Wang, Fanchieh Yee, Jintao Zhang, Matthew Ziegler, Ching Zhou, Moriyoshi Ohara, Pong-Fei Lu, Brian Curran, Sunil Shukla, Vijayalakshmi Srinivasan, Leland Chang, and Kailash Gopalakrishnan, Proceedings of the IEEE, 2020
• DeepTools: Compiler and Execution Runtime Extensions for RaPiD AI Accelerator [link]
  Swagath Venkataramani, Jungwook Choi, Vijayalakshmi Srinivasan, Wei Wang, Jintao Zhang, Marcel Schaal, Mauricio Serrano, Kazuaki Ishizaki, Hiroshi Inoue, Eri Ogawa, Moriyoshi Ohara, Leland Chang, and Kailash Gopalakrishnan, IEEE Micro, 2019
• Enjoy Data Processing [I] : Java Language　（講座:データ処理のためのプログラミング言語 [I] -Java言語編-） [pdf]
  Kazuaki Ishizaki
  Journal of the Institute of Electronics, information and communication Engineering, Vol. 102 No. 6, pp. 583-587, 2019 (in Japanese) , copyright©2019 IEICE
• Code Optimization in HPF Compiler for SP2 [link]
  Hideaki Komatsu, Toshio Suganuma, Takeshi Ogasawara, Kazuaki Ishizaki, and Osamu Gohda
  Journal of Information Processing Society of Japan, Vol. 38 No. 2, pp. 100-104, 1997 (in Japanese) 

Back to top

• A Study of Devirtualization Techniques for a Java Just-In-Time Compiler
  Kazuaki Ishizaki, Motohiro Kawahito, Toshiaki Yasue, Hideaki Komatsu, and Toshio Nakatani
  IBM Research Report, RT0352, 2000
• Optimizations for Reducing the Runtime Overhead of the Java Language in a Just-In-Time Compiler
  Kazuaki Ishizaki, Motohiro Kawahito, Toshiaki Yasue, Mikio Takeuchi, Takeshi Ogasawara, Toshio Suganuma, Tamiya Onodera, Hideaki Komatsu, and Toshio Nakatani
  IBM Research Report, RT0285, 1998

Back to top