Index

Parallel Architectures

In the processor section we learned about instruction-level parallelism and how there are limits to how much parallelism there is in our applications. However, we can overcome this, by leveraging multiple processors.

In this section, we’re going to learn about how to combine multiple independent instruction streams executing on different hardware to improve performance.

Parallel system’s performance

• Amdahl’s law revisited

Parallel Architectures and Programming

• Kinds of parallel architectures
• Shared memory architectures

Parallel Memory Systems

• Cache coherence and memory consistency
  • Cache coherence
  • Memory consistency

Accelerator architecture

• GPUs
• Other accelerators

This time, both 154B and 201A are going to do the same gem5 assignment!

154B/201A Assignment 5 DUE 3/15

---