Course Material
| Chapter | Topic | Presentation | Lecture | |
| 0 | Course Overview | Lecture Notes | ||
| 1 | Introduction to Performance | 4 slides per page | 1 slide per page | Lecture Notes |
| 2 | Instruction Set Architecture (ISA) | 4 slides per page | 1 slide per page | Lecture Notes |
| 3 | Performance Theory | 4 slides per page | 1 slide per page | Lecture Notes |
| Summary of Definitions and Formulas | ||||
| 4 | From CISC to RISC | 4 slides per page | 1 slide per page | Lecture Notes |
| 5 | DLX Architecture | 4 slides per page | 1 slide per page | Lecture Notes |
| 6 | Speeding-up DLX | 4 slides per page | 1 slide per page | Lecture Notes |
| 7 | Computer Arithmetic | 4 slides per page | 1 slide per page | Lecture Notes |
| 8 | Memory and I/O Organization | 4 slides per page | 1 slide per page | Lecture Notes |
| 9 | Advanced Architectures | 4 slides per page | 1 slide per page | Lecture Notes |
| 10 | Real Life RISC | 4 slides per page | 1 slide per page | Lecture Notes |
![]() |
All slides in one file | 1 slide per page | Lecture Notes | |
Exercises
![]() |
Exercise 1 | |
![]() |
Exercise 2 | |
![]() |
Exercise 3 | |
![]() |
Exercise 4 | |
![]() |
Exercise 5 | |
![]() |
Exercise 6 | |
![]() |
Exercise 7 | |
![]() |
Exercise 8 | |
![]() |
Exercise 9 | |
![]() |
Exercise 10 | |
Final Paper / עבודה מסכמת
90% of the final grade will be based on a final paper, submitted in two stages.
In the first stage, you will write a simple program in C and compile it to RISC assembly language.
In the second stage, you will analyze and improve the run time based on methods learned in the course.
The paper will be submitted in groups of two or three.
The passing grade is 55. As in any final paper, the grade will not only be based on writing correct statements, but also on the overall quality of the submission as an academic paper.
The submission dates is are given in the syllabus (after the exam period of spring semester). There is no second submission period for final papers.
Additional details will be available on the course website.
Notice that there are two versions of the Final Paper Assignment, one for submission by two students and one for submission by three students (including 50% additional work).
Nevi'im: Assignment for groups of two Assignment for groups of three
Strauss: Assignment for groups of two Assignment for groups of three
For Part 1 of the Assignment, save the following Windows programs in the same folder as your C program
| DLL for Cross-Compiler | MIPS32 Cross-Compiler for Windows |
and follow the instructions in the Final Paper Assignment.
For Part 2 of the Assignment, the document MIPS32 for DLX Programmers contains detailed information on the MIPS32 CPU.
Additional information is found in the Final Paper Assignment.
Further Reading
![]() |
Introduction to Performance |
| Intel 80186 Data Sheet, pages 27 - 32 | |
| Standard Performance Evaluation Corporation (SPEC) | |
![]() |
Instruction Set Architecture (ISA) |
| Hennessy & Patterson, Computer Architecture, chapter 2 | |
![]() |
Performance Theory |
| Hennessy & Patterson, Computer Architecture, chapter 2 | |
![]() |
From CISC to RISC |
| Hennessy & Patterson, Computer Architecture, chapter 2 | |
![]() |
DLX Architecture |
| Hennessy & Patterson, Computer Architecture, chapter 2 | |
| Hennessy & Patterson, Computer Architecture, chapter 3 | |
![]() |
Speeding-up DLX |
| Hennessy & Patterson, Computer Architecture, chapter 3 | |
![]() |
Memory Organization |
| Hennessy & Patterson, Computer Architecture, chapter 5 | |
![]() |
I/O System Performance |
| Hennessy & Patterson, Computer Architecture, chapter 7 | |
