Parallelism vs. Concurrency for Engineers
Correctly diagnose whether a system's bottleneck is I/O-bound or CPU-bound, and choose between concurrent structuring (async, goroutines, event loops) and parallel execution (multiprocessing, multi-core, GPU) accordingly.
4 sections ยท 8 lessons
Course outline
Definitions and the Core Distinction
- Concurrency is structure, parallelism is execution
- Why the two get conflated
When Concurrency Wins
- I/O-bound workloads and event loops
- Concurrency without multiple cores
When Parallelism Wins
- CPU-bound bulk computation
- The Python GIL: why threads aren't enough
Combining Both
- Go's model: concurrent design, parallel execution
- A decision framework