2110714 Digital Systems 

1st semester 2018 (Aug)
classroom :  Thursday  13:00-16:00, Eng 3,  room 403

old lecture  2017  2016 


Today digital technology permeates every corner of our society. There are three big trends that shape our future. The first one is the coming of Artificial Intelligence.  The second one is the rise of Automation.  The last one the revolution of technology to transfer money.  These advancements rely on the power of computing.  This class is divided into two parts: the first one discusses fundamental of performance of the modern processor, the second part discusses advanced in computer architecture, emphasized on Graphics Processing Unit.


Fundamental of performance
Performance measure
Cache memory
Virtual memory
Advanced topics
Graphics Processing Unit

How we learn

We learn by doing both hardware simulation and software.  The assignments will be given out periodically throughout the semester. We will also do in class running demo and programming.  Please take a notebook with you  into the class. The final exam is "open book" and has an element of system design plus some "pseudo" programming.


  16 Aug 2018   start of the class
  27 Nov 2018   the date of final project presentation is Friday 14 Dec 2018, 1pm at ISL Eng Building 4.
  . . .


30%   homework
40%   final (open book)
30%   project


Performance  Chapter 1
Memory Hierarchy  Chapter 2
Instruction level paralellism  Chapter 3
Data parallelism  Chapter 4
Interrupt and its programming
Low power programming (with simulated IoT board)
     example interrupt mode program:   (in Rz)  demo-rz.txt
Operating systems and task scheduling basic
OS concepts
Multi-task OS  (MOS)      semaphore video
Graphic Processing Unit (GPU)
Introduction to CUDA programming

Scientific Computing
Quantum Computers
final exam
project presentation

Next week

Please prepare your machine for next week in class activities

. . .


Tools for interrupt programming   iot-rz-17.zip
GPU simulator  npsim4-1.zip


8 Nov 2018   Read Scientific computing, choose one of two reports on exascale computing.  Write about one of these topics:
a)  applications of exascale computers. (can be either in terms of national policy, the world progress of exascale ... )
b)  the architecture of exascale. (the advancement of exascale technology, the research done in the area etc...)
You must hand-in 4-5 pages of report by next week.
. . .


This is equivalent to final exam.  You are required to write  an in-dept report on the topic of GPU (mostly for GPGPU).  The topics can be varied according to your interest.  The topics include: using GPU for parallel processing, GPU for special purpose computing, the latest architecture of GPU, GPU programming etc.  I expect it to be about one week work.  You are asked to present your work to the class.  The presentation length is 8 minutes.  The meeting is around one week "after" the final exam period.  Meeting will take place at my research lab on 20th floor, Engineering building 4.  The exact date is to be confirmed.


(27 Nov 2018)

There will be no final exam.
All the work credited to the score are homework and a final project.
You already submit all your homework.
The final project is "GPU related" work (it centered around making use of CUDA programming).
You are required to submit the report of your final project on 14 December.  (I will need to grade and submit it by that day).
We will meet in a class (at my research lab on 20th floor, Eng Build 4) on 14 December (Friday) at 1pm. Everyone will present their work. The time slot is 8 min.
If you cannot make the final presentation on that day, you can arrange your presentation to be before 14 Dec.  Please contact me.

Additional information

Reference textbook:  Computer Architecture: A Quantitative Approach 5th Edition by John L. Hennessy, David A. Patterson, Morgan Kaufmann, 2011.  link to Amazon

last update 27 Nov 2018