Course Description
Provides a strong foundation to understand computer system architecture from a quantitative perspective and apply it to modern computer designs. Students will utilize hardware and software modeling to implement concepts presented in the class. The class will include both individual and group components.
Additional Requirements for Graduate Students:
The graduate student will prepare a class presentation and term report analyzing a current advancement in the field either through patent search and/or academic research papers. The topic is chosen by the student but should be focused on a topic related to the class. The goal is to quantitatively evaluate, assess, and communicate the findings to the class.
Athena Title
Computer Architecture
Prerequisite
ECSE 2920 or ECSE 2920H
Semester Course Offered
Offered every year.
Grading System
A - F (Traditional)
Course Objectives
By the end of this course, a successful student will be able to: - Describe computer architecture concepts related to processor, memory, and systems, and explain how these interact. - Apply these concepts to computer architecture design problems by balancing application constraints with technology constraints: power, execution time, quality, area, and cost. - Evaluate various alternatives via quantitative and qualitative assessments to determine the superiority of design and approaches. - Demonstrate the ability to design, test, and verify designs using hardware and/or software models at the systems and register level. - Create new designs based on defined requirements and objectives and test these designs with appropriate strategies. - Write technical reports describing the created designs, trade-offs, verification and test strategies, and evaluation of current and future techniques to determine best practices. Graduate students Along with the above, the student should: - Independently identify and research advancement in the field. - Quantitatively assess and evaluate the advancement using engineering principles and metrics from the class. - Effectively communicate, demonstrate, and teach the advancement through oral and written means.
Topical Outline
- Foundations of processors - ISA (Instruction Set Architecture) - FSM review - HDL review - Single-cycle - Multi-cycle - Pipelined - Hazards - Performance - RISC vs. CISC overview - CPU vs. GPU - Memories - Memory technology - Memory topology - Caches > Direct > Associative > WT vs. WB - Protection - Virtualization - Performance - Advanced processor topics - Superscalar - OOE - Register renaming - Memory - Prediction - Speculative execution - Multi threads - VLIW - Future topics - Products in the news - New techniques in the news
Syllabus
Public CV