---
title: "Parallel processing and pipelining"
subject: "Computer Organization and Architecture"
chapter: 4
key: "semester_5/computer_organization_and_architecture:parallel-processing"
source: https://www.bctengineer.com/q/semester_5/computer_organization_and_architecture/parallel-processing
site: BCT Engineering
---

# Parallel processing and pipelining

Computer Organization and Architecture · Chapter 4: Pipeline & Vector Processing

- Asked in: 2071 Shrawan, 4 marks
- Video: [Parallel Processing in Computer Organization Architecture || Pipelining || Flynn classification comp](https://www.youtube.com/watch?v=0l_KkiaQECk) — Sudhakar Atchala
- Answer checked by Claude Opus 5.5 (high effort) on 2026-10-14
- Page: https://www.bctengineer.com/q/semester_5/computer_organization_and_architecture/parallel-processing
- Whole chapter: https://www.bctengineer.com/md/semester_5/computer_organization_and_architecture/chapter-4

## Question

Discuss parallel processing. How can parallel processing be achieved in pipelining?

## Short answer

Parallel processing uses multiple units to perform tasks simultaneously. Pipelining achieves temporal parallelism by overlapping instruction stages; for k stages and n instructions, speedup approaches k, giving one result per clock after the pipe fills.

## Answer

Parallel processing is the use of multiple processing units or overlapped operations to perform several tasks at the same time, so that the system's throughput and speed increase. Instead of processing in sequence, the work is divided into parts that are done together.

Levels: job or program level, task level, instruction level (pipelining), arithmetic and bit level.

#### Parallelism achieved in pipelining
A pipeline is a form of parallel processing in time (temporal parallelism). The instruction cycle is divided into segments (FI, DA, FO, EX) that work simultaneously on different instructions.
- Several instructions are in different phases at one time, so one instruction is completed per clock once the pipe fills.
- The time for n instructions is (k + n - 1) tp instead of n·k·tp, so the speedup approaches k.
- More parallelism comes from several pipelines (superscalar), multiple functional units and vector pipelines.

```text
 Clock:   1  2  3  4  5  6
 I1       FI DA FO EX
 I2          FI DA FO EX
 I3             FI DA FO EX
```
Other ways: multiple functional units, SIMD array processors and multiprocessors.

## Check yourself

The answer's key points as multiple-choice questions, with the correct option and why.

1. What is the aim of parallel processing?
   A. To process tasks strictly in sequence
   B. To perform several tasks at the same time and increase throughput
   C. To reduce the number of instructions
   D. To reduce memory size
   Answer: B. Work is divided into parts done together so system speed and throughput rise. (Key point: Meaning of parallel processing)

2. Why is pipelining called parallelism in time (temporal parallelism)?
   A. Segments work simultaneously on different instructions
   B. It uses many identical processors
   C. All instructions use the same data
   D. It runs without a clock
   Answer: A. FI, DA, FO and EX work at the same time on different instructions. (Key point: Pipelining as parallelism)

3. Time for n instructions in a k-segment pipeline compared with no pipeline is:
   A. (k + n - 1) tp instead of n·k·tp
   B. n·tp instead of k·tp
   C. k·tp instead of n·tp
   D. (n - k) tp instead of n·k·tp
   Answer: A. Overlap gives speedup approaching k. (Key point: Time for n instructions)

4. Which gives more parallelism beyond a single pipeline?
   A. Superscalar with several pipelines or multiple functional units
   B. Using a single functional unit
   C. Removing registers
   D. Using a longer clock
   Answer: A. Several pipelines, multiple functional units, SIMD arrays and multiprocessors add parallelism. (Key point: More parallelism)

## Related in Computer Organization and Architecture

- [Vector processing and pipelining performance](https://www.bctengineer.com/question.md?key=semester_5%2Fcomputer_organization_and_architecture%3Avector-pipeline-performance) — Computer Organization and Architecture, chapter 4, asked 2×

## Related in other subjects

- [Parallel processing and pipeline speedup](https://www.bctengineer.com/question.md?key=semester_4%2Fmicroprocessor%3Aparallel-processing-pipeline-speedup) — Microprocessor, chapter 6