Parallel and
Distributed
Computing
We will discuss:
Hardware Architecture
Flynn’s Taxonomy
Types of Flynn’s Taxonomy
What is hardware architecture?
• In Parallel and Distributed Computing Hardware Architecture refers
to Identify system’s physical components and their interrelationship.
• Hardware Architecture, often called hardware design model, it
allows hardware designers to understand how their components fit
into a system architecture and provide to software components
designers important information needed for software
development and integration.
• Architecture:
• Basically Architecture is a set of rules and methods that describe
the functionality, organization, and implementation of computer.
• Architecture is relate with balancing the performance, efficiency,
cost, and reliability of computer system.
Diagram of Hardware Architecture
Flynn’s Taxonomy:
Michaeal j. Flynn (born May 20, 1934).
American professor at Stanford university.
Flynn, M. J. Proposed Flynn’s Taxonomy (1966).
Method of classifying digital computer architectures.
Very high-speed computing systems.
Types of Flynn’s taxonomy
1.Single instruction stream and Single Data Stream (SISD)
2.Single instruction stream and Multiple Data Stream (SIMD)
3. Multiple instruction streams, and Single Data Stream (MISD)
4. Multiple instruction streams, Multiple Data Stream (MIMD)
Instruction stream Vs Data stream
Instruction stream:
The sequence of instructions from memory to control unit.
Data stream:
The sequence of data from memory to control unit.
1. SISD
SISD (Single Instruction,
Single Data) This is a sequential
model where a single processor
executes one instruction at a time
on a single data stream. A
standard personal computer with
a single CPU is a good example.
2. SIMD
SIMD (Single Instruction, Multiple
Data)
In this model, a single instruction is
executed on multiple data streams
simultaneously. Graphics processing
units (GPUs) often utilize this
architecture.
3. MISD
MISD (Multiple Instruction,
Single Data) This model involves
multiple processors executing
different instructions on the same
data stream. While theoretically
possible, it's not widely used in
practice.
4. MIMD
MIMD (Multiple Instruction, Multiple
Data)
This is the most common parallel
architecture, where multiple processors
independently execute different
instructions on different data
streams. Modern multi-core processors
and supercomputers often employ this
architecture.
