Infrastructure & Compute
Parallelism
Performing multiple computations at the same time to speed up processing.
Example
Different GPUs process separate batches at the same time.
Why it matters
It can shorten work that divides into suitable simultaneous tasks.
You'll hear it
“These calculations can run together, but that last step must wait.”
Can you control it?
Set by the team
The people building or running the AI choose this setup. An everyday user generally needs their help to change how this part works.
Common questions
- Can coordination cost more than the time saved?
- Yes. Small tasks may not benefit enough to justify splitting them.
- Does parallel mean every task finishes at the same time?
- No. Different pieces can have different durations.
- Can some work remain necessarily sequential?
- Yes. A later step may require an earlier result.