Course:
Course Program:
Computer architecture is one of the oldest but also most research-active fields in the domain of electrical and computer engineering. Certain factors have always transformed the computer architecture. Today, these factors are data explosion and artificial intelligence. In this course, while learning the basic terminologies and methods in computer architecture, we will also mention about hot research areas. In this course, we will study the following topics:
- Current State of the Art. Research trends.
- Definitions. Evaluation metrics.
- Instruction Set Architecture
- Memory Hierarchy
- Parallelism: instruction-, data-, thread-level
- Interconnects
- Domain Specific Architectures
Textbook:
J. L. Hennessy and D. A. Patterson, Computer Architecture: A Quantitative Approach, Sixth Edition, Morgan Kaufmann, 2019.
Reference Books:
Recent papers from the literature
Grading:
- Active attendance: 10 pts
- Homeworks: 60 pts (for each submission, late is penalized by 20% per day (including weekends))
- Term paper: 30 pts (for each of the following item, late is penalized by 100% per minute)
- Topic presentation: 5 pts (due date: May 5th, lecture time)
- Interim presentation: 10 pts (due date: June 9th, lecture time)
- Term paper submission: 15 pts (due date: Final date and time)
Notes:
- Academic honesty is required. Cheating (both ways), plagiarism (including mosaic plagiarism), taking materials from internet or from friends and unfair use are fully forbidden.
- Catalog grading is applied:
- Active attendance requires the student to be physically present in the class, present his/her work, ask and answer inspiring questions
- Homeworks can be in the form of pen-paper, coding, simulation, research.
- In term paper, the student will start with a recent paper from one of these conferences: MICRO, ISCA, ASPLOS, ASAP, NOCS, PACT, ARCS, CARRV. For other conferences or publications such as journals and magazines, you need to consult the instructor. After topic selection, a further analysis needs to be carried out to understand the source of problem and the possible solutions.