2017 Large Scale Computing Systems

Font size  SML

Register update notification mail Add to favorite lecture list
Academic unit or major
Graduate major in Information and Communications Engineering
Sugino Nobuhiko 
Course component(s)
Day/Period(Room No.)
Course number
Academic year
Offered quarter
Syllabus updated
Lecture notes updated
Language used
Access Index

Course description and aims

This lecture forcuses recent high-performance processor architectures such like multi-core processors and GPUs, and covers various techniques (programming techniques, code optimization methods, and tuning schemes) for higher peformance computing on these processors.
This course has two aims. The first is to teach students to understand various high-performance processor architectures. The other is to teach students to understand software techniques including programming techinques, code optimization techniques, turing techniques, and etc., in order to achieve higher performance
on these high-performance processor architectures.

Student learning outcomes

At the end of this course, students will be able to understand
- Techniques for High-Performance Computing on Multi-core processos
- Techniques for High-Performance Computing on GPU


Parallel Computing, High-Performance Computing, Multi-Core Processors, GPU

Competencies that will be developed

Intercultural skills Communication skills Specialist skills Critical thinking skills Practical and/or problem-solving skills
- - - -

Class flow

1) At the beginning of each class, the previous class are reviewed.
2) Towards the end of class, students are given assignments related to what is taught on that day to solve.
3) Before coming to class, students should read the course schedule and check what topics will be covered. Required learning should be completed outside of the classroom for preparation and review purposes.

Course schedule/Required learning

  Course schedule Required learning
Class 1 Parallel computing on Multi-core processors Understand outline of parallel computing on multi-core processors
Class 2 Techniques for Higher Performance Computing on Multi-core processors Understand major techniques for higher performance computing on multi-core processors
Class 3 GPU architectures and GPGPU Learn GPU architectures and understand GPGPU.
Class 4 Parallel / Vector Processing with Thread Programming Model. Understand parallel / vector processing with thread programming model.
Class 5 Memory Architectures of GPUs and their Data Transfer Understand memory architectures of GPUs and their data transfer
Class 6 Quatitative Evaluation of Program Codes Understand major quatitative evaluation measures of program codes
Class 7 Turing Techniques for Higher Performance Computing Learn turing techniques for higher performance computing on GPUs.
Class 8 Various Tuning Techniques Learn various tuning techniques for GPU programs.


None required.

Reference books, course materials, etc.

Course materials are provided during class(can be found on OCW-i).

Assessment criteria and methods

1) Students will be assessed on their understanding of parallel computing on multi-core processors and GPUs, and techniques for high-performance computing on them.
2) Students’ course scores are based on final exams (80%) and assignments (20%).
3) Full attendance is compulsory.
4) The instructor may fail a student if he/she repeatedly comes to class late or resubmits assignment reports too often.

Related courses

  • ICT.I415 : VLSI System Design
  • ICT.H416 : Statistical Theories for Brain and Parallel Computing
  • ICT.H507 : Virtual Reality and Interaction
  • ICT.I308 : Computer Architecture (ICT)
  • ICT.I303 : Integrated Circuit Design
  • ICT.I216 : Computer Logic Design (ICT)

Prerequisites (i.e., required knowledge, skills, courses, etc.)

No prerequisites are necessary, but students are recommended to have
the basic knowledge on hardware and software of computer systems.

Page Top