One cook or several
One cook can prepare several dishes by switching between them: concurrency. Several cooks preparing dishes at once is parallelism.
A single processor can handle concurrent tasks by switching between them. Multiple processor cores or machines can work in parallel.
Overlap creates coordination
Concurrency lets a program work on one task while another waits for a network response or a file to load. Parallelism can finish some jobs faster by splitting the work. When tasks share data or resources, both approaches need coordination to avoid race conditions. Extra workers can also compete for resources, so doubling their number does not always halve the time.