| |
Apr 19, 2024
|
|
|
|
|
CE 4930 - Computer Architecture II3 lecture hours 0 lab hours 3 credits
Course Description
Modern microprocessor architectures extend pipelined micro-architecture in a number of ways in order to exploit instruction-level parallelism (ILP) and thread-level parallelism (TLP). Deep pipelines, superscalar pipelines, out-of-order instruction execution, instruction re-ordering and speculative execution are example techniques exploiting ILP. Similarly, multiprocessor techniques such as maintaining a coherent shared memory among multiple cores are examples that exploit thread-level parallelism. These examples challenge the fundamental architectural concept of single-instruction per clock-cycle and result in circuits that improve performance and enrich the user experience. This course explores these topics through lecture, in-class problems, reading assignments, and homework. (prereq: CE 1921 )
Course Learning Outcomes
Upon successful completion of this course, the student will be able to:
- Describe how deep pipelines exploit instruction level parallelism and increase clock rate
- Describe how superscalar processors exploit instruction level parallelism to increase IPC
- Describe how out-of-order execution improves performance in superscalar processors
- Describe how speculative execution improves performance in microprocessor pipelines
- Compare and contrast static and dynamic speculative execution techniques
- Describe how multiprocessors exploit instruction and thread level parallelism
- Discuss classic microprocessor case studies such as the MIPS R4000, Intel Pentium, Motorola 88110, Intel Pentium Pro, and IBM Cell multiprocessor
Prerequisites by Topic
Course Topics
Coordinator
Dr. Russell Meier
Add to Portfolio (opens a new window)
|
|
Skip to Main Content
