Parallel Computing
Parallel computers split work across many processors at once.
About this topic
Parallel computing means splitting a task into parts that run at the same time on multiple processors or cores, instead of one step after another. Modern phones and laptops have several cores, GPUs have thousands of simple ones, and supercomputers link many thousands of processors. It lets us solve bigger problems faster.
Not every task can be shared. The speed-up depends on how much of the work can be done in parallel, a limit described by Amdahl's law: if 20 percent of a job must run serially, even infinite processors give at most a five times speed-up. Parallel programs must also deal with communication, sharing data safely, and problems like race conditions, where the result depends on timing, and deadlock, where tasks wait for each other forever.
Painting a large wall is a good illustration: ten painters finish much faster than one, but only if they have enough space and do not wait for each other. Parallelism is used in graphics, weather forecasts, scientific simulations, video editing and training AI models. A common misconception is that doubling processors always doubles speed. Another is confusing concurrency (managing many tasks) with true parallelism (running them simultaneously).
Key ideas
- Multicore CPUs and GPUs
- Some tasks split easily
- Speedup is limited by serial parts
Quick check
Use cases: where Parallel Computing helps
- Homework: review the key idea behind Parallel Computing before you start a problem set.
- Exam revision: use the quick check to test whether you can recall Parallel Computing without notes.
- Visual learning: play with the interactive lab to see how Parallel Computing changes when you adjust its inputs.
- Teaching a friend: explain Parallel Computing in your own words, then compare your version with the key ideas.
Computing ideas run phones, search engines, games and the wider internet.
Frequently asked questions
What is Parallel Computing?
Parallel computers split work across many processors at once.
What will I learn in Parallel Computing?
Multicore CPUs and GPUs. Some tasks split easily. Speedup is limited by serial parts.
Is Parallel Computing part of Computer Science?
Yes. It is one of 80 Computer Science topics on LearnViz HighSchoolEdu, with 3 key ideas, a live animation and a quick check.
Does the Parallel Computing lab work on a phone or tablet?
Yes. The animation resizes to fit your screen and supports touch, so you can drag, tap and slide with your finger.
How should I study Parallel Computing?
Read the summary and key ideas first, play with the interactive lab to see the idea in motion, then answer the quick check. If you get a question wrong, go back to the key idea it tests.
Do I need an account, and is my progress saved?
No account is needed. Marking a topic as learned is saved in your own browser, so it stays on that device and is not sent to a server.