Scribd
Upload
7 views7 pages

Assignment 1

s

Uploaded by

zohaibsoomro100
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
7 views7 pages

Assignment 1

s

Uploaded by

zohaibsoomro100
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 7

Pakistan Studies

PROJECT ON MAULANA MAUDUDI

NAME: MUHAMMAD ZOHAIB


CLASS: 7th Semester, Section-C, BSCS
STU ID: CSC-20F-132
TEACHER’S NAME: SIR ASIF MAGSI
4 Architectures of FLYNN’S classification
Flynn’s classification is a well known taxonomy for categorizing
computer architectures based on the number of instruction streams
and data steams. There are four main architectures in Flynn’s
classification.

SISD (Single Instruction, Single Data)


 In SISD architecture, a single stream of instructions is processed
on a single stream of data.
 This is the traditional von Neumann architecture, where a central
processing unit (CPU) executes instructions one at a time on a
single piece of data.
 Most general-purpose computers, including personal computers
and workstations, are SISD machines.

Example:
The classic example of a SISD architecture is the original IBM 360/91
mainframe computer, introduced in 1967. It had a single central
processing unit (CPU) that executed instructions sequentially on a single
stream of data.

SIMD (Single Instruction, Multiple Data):


 In SIMD architecture, a single instruction is used to perform the
same operation on multiple pieces of data in parallel.
 SIMD computers have multiple processing elements (typically
vector processors) that can operate on different data elements
simultaneously.
 This architecture is commonly used for tasks that can benefit from
parallelism, such as graphics processing units (GPUs) and some
supercomputers.

Example:
The Connection Machine, a series of supercomputers developed by
Thinking Machines Corporation in the 1980s and 1990s, is an example
of SIMD architecture. These machines used multiple processing
elements to execute the same instruction on different data elements in
parallel, making them suitable for tasks like scientific simulations and
artificial intelligence.

MISD (Multiple Instruction, Single Data):


 In MISD architecture, multiple instruction streams are used to
operate on a single stream of data.
 This architecture is relatively rare and is not commonly found in
general-purpose computing. It has been studied and used in
specialized applications like fault-tolerant systems and some signal
processing scenarios.

Example:
MISD architecture is not commonly used, and practical examples are
scarce. One theoretical example could be a system for fault tolerance or
error correction in which multiple instruction streams are applied to the
same data stream to detect and correct errors. However, practical
implementations are limited.

MIMD (Multiple Instruction, Multiple Data):


 In MIMD architecture, multiple instruction streams can operate on
multiple data streams independently and concurrently.
 MIMD is the most common and versatile architecture, as it allows
for true parallelism and is widely used in multiprocessor systems,
clusters, and modern multi-core processors.
 Each processing unit in a MIMD system can execute its own set of
instructions on its own data, making it suitable for a wide range of
applications, including general-purpose computing, scientific
simulations, and parallel computing tasks.

EXAMPLE:
The Cray X-MP, a supercomputer developed by Cray Research in the
1980s, is an example of a MIMD architecture. It had multiple
processors, each capable of executing its own set of instructions on its
own data, allowing for parallel processing of various tasks in scientific
and engineering applications.
4 Architectures of FENG’S classification
Tse-yun Feng introduced a categorization system for parallel processing
systems, which hinges on the word length and the number of bits in
each word. This classification is grounded in the parallelism exhibited by
both bits and words. As a result, the processing of word bits can occur
either in parallel or sequentially. Likewise, the processing of words can
be conducted in a parallel or sequential manner. This classification
yields four distinct categories.

Word Serial Bit Serial (WSBS):


In this scenario, individual bits of a chosen word are handled
sequentially, leading to a process that is performed serially, resulting in
the maximum processing time being needed.

Example:
An example of WSBS could be serially checking and processing
individual bits of a packet's header to make routing decisions, which
may be slower but can simplify the network's logic and reduce
hardware complexity.

Word Serial Bit Parallel (WSBP):


In this situation, all the bits within a chosen word are processed
concurrently, albeit one word at a time. Therefore, "bit parallel" refers
to the simultaneous selection of all the bits within a word, making it a
form of processing that is somewhat parallel in nature.
EXAMPLE:
An example of WSBP in networking is decoding multiple Huffman-
encoded data streams simultaneously, with each data word's bits being
processed concurrently to improve data compression and
decompression efficiency.

Word Parallel Bit Serial (WPBS):


In this scenario, a single designated bit is processed at a time from a set
of specified words. Therefore, "word parallel" denotes the selection of
multiple words. WSBP can be described as a form of parallelism that
operates row-wise, while WPBS can be likened to a type of parallelism
that operates column-wise.

EXAMPLE:
In error correction, WPBS can be used to check and correct individual
bits across multiple data streams simultaneously to improve data
reliability.

Word Parallel Bit Parallel (WPBP):


In this situation, all the bits from the specified words are processed
simultaneously, achieving the highest degree of parallelism and, as a
result, the shortest execution time. The processors categorized
according to Feng's Classification can be visually represented by plotting
the number of bits against the number of words processed in parallel.
Example:
In a network switch, WPBP can be used to process multiple bits of an
incoming data packet's header at the same time, allowing for faster and
more efficient routing decisions.

You might also like