Modern computers consist of digital circuits implemented on silicon wafers. However, humans program with textual, high-level programming languages such as C++ and Python, and it is not obvious how abstract computing models that humans perceive are actually executed on hardware. In practice, there are many layers involved in the execution, and for this particular course, first we study language abstraction at the lowest level, namely machine languages, and then how hardware can be constructed to interpret and execute such machine languages. In the lab course we will also learn how to construct physical computers.
The goal of this course is to learn the hardware architectural concepts of computers, how they execute the programs in principle. Also, we will cover recent advances in computer architectures, especially techniques to attain speed and execution efficiency, in order to attain insights into how an efficient computing infrastructure can be constructed.
We will first be learning the lowest level programming language, namely machine language, and will then learn how hardware components such as ALUs and memory devices are designed with digital circuits. This will allow students to come to a complete understanding of the abstraction layers of computers based on the von Neumann architecture.
Computer architecture, machine language, addressing, von Neumann machine, digital circuits, adders, expressing values with binary numbers, ALU, combinatorial logic, truth table, finite state logic, data path, single cycle execution, multi cycle execution, finite state machine / automaton, microprograming, pipelining, storage / memory, DRAM, memory hierarchy, cache, paging. MMU, I/O, data bus, PCI. hard disk, flash memory, performance, Amdahl’s law.
✔ Specialist skills | Intercultural skills | Communication skills | Critical thinking skills | ✔ Practical and/or problem-solving skills |
For each lecture class a set of PowerPoint slides will be distributed prior to the class, and it is expected that the student will study the material prior to the class. The lecture itself will proceed interactively. The lab course will have an assignment every one or two classes, and a report is to be submitted for each assignment. The grade will be determined based on the combination of the score of the final exam and the submitted reports.
Course schedule | Required learning | |
---|---|---|
Class 1 | Introduction to computer systems, computing history | Slides to be studied prior to class |
Class 2 | Introduction to computer architecture, assembly / machine language (1) arithmetic and logical instructions | Slides to be studied prior to class |
Class 3 | Assembly / machine language (2): load/store instructions, control instructions | Slides to be studied prior to class |
Class 4 | Assembly / machine language (3): implementing procedures and functions, different instruction set architectures | Slides to be studied prior to class |
Class 5 | Performance of computers (1): performance modeling | Slides to be studied prior to class |
Class 6 | Performance of computers (2): various performance metrics | Slides to be studied prior to class |
Class 7 | Introduction to digital circuits (1): combinatorial logic, truth table, and disjunctive normal form | Slides to be studied prior to class |
Class 8 | Introduction to digital circuits (2): Expressing values with binary numbers, implementing ALUs | Slides to be studied prior to class |
Class 9 | Introduction to digital circuits (3): various finite state logics and their design | Slides to be studied prior to class |
Class 10 | CPU architecture (1): single cycle implementation | Slides to be studied prior to class |
Class 11 | CPU architecture (2): multi cycle implementation | Slides to be studied prior to class |
Class 12 | CPU acceleration: pipelining and superscalar | Slides to be studied prior to class |
Class 13 | Memory hierarchy, cache memory, and virtual memory | Slides to be studied prior to class |
Class 14 | various I/O and communication networks | Slides to be studied prior to class |
Class 15 | hardware security technologies | Slides to be studied prior to class |
Not Specified
Not specified
The grade will be determined based on the combination of the score of the final exam and the submitted reports (50% each).
Basic knowledge of automata theory, programming, and Boolean logic