overview.txt

This course aims to provide a strong foundation for students to
understand modern computer system architecture and to apply these
insights and principles to future computer designs. The course is
structured around the three primary building blocks of general-purpose
computing systems: processors, memories, and networks.

The first half of the course focuses on the fundamentals of each building
block. Topics include instruction set architecture; single-cycle
processors; hardwired vs. microcoded FSM processors; pipelined
processors; direct-mapped vs. associative caches; pipelined caches;
network topology, routing, and flow control; and integrating processors,
memories, and networks. The second half of the course delves into more
advanced techniques and will enable students to understand how these
three building blocks can be integrated to build a modern shared-memory
multicore system. Topics include superscalar execution, out-of-order
execution, register renaming, memory disambiguation, branch prediction,
and speculative execution; multithreaded, VLIW, and SIMD processors;
non-blocking cache memories; memory protection, translation, and
virtualization; and memory synchronization, consistency, and coherence.
Students will learn how to evaluate design decisions in the context of
past, current, and future application requirements and technology
constraints.

A significant project is decomposed into five lab assignments. Throughout
the semester, students will gradually design, implement, test, and
evaluate a complete multicore system capable of running simple parallel
applications at the register-transfer level.