2. Contents
What is Parallel Processing
Why parallel processing
How parallel processing divide workload
Classification Parallel Processor Architectures Classification Parallel Processor Architectures
Advantages of Parallel processing
3. Parallel processing
In Parallel processing a single program is run
by multiple processor
Parallel processing is a form of processing in
which many instructions are carried out
simultaneously operating on the principle
which many instructions are carried out
simultaneously operating on the principle
that large problems can often be divided into
smaller ones, which are then solved
concurrently (in parallel)
4. Parallel processing
Is the method of evenly distributing computer
processes between two or more computer
processors
This requires a computer with two or more This requires a computer with two or more
processors installed and enabled
It also requires an operating system capable
of supporting two or more processors, and
software programs capable of evenly
distributing processes between them
5. Why parallel processing
With the increased use of computers in every
sphere of human activity, computer scientists
are faced with crucial issues today:
Processing has to be done faster like never
beforebefore
Larger or complex computation problems
need to be solved
Early computers one ALU that perform one
operation at a time mean’s Slow processing
6. Why parallel processing
Increasing the number of transistors as per
Moore’s Law isn’t a solution, as it also increases
the frequency scaling and power consumption
Power consumption has been a major issue Power consumption has been a major issue
recently, as it causes a problem of processor
heating
7. Why Parallel Processing?
Traditional computers often are not able to
meet performance needs in many applications:
Simulation of large complex systems in physics,
economy, biology.... etceconomy, biology.... etc
Distributed data base with search function
Computer aided design
Visualization and multimedia
8. Why Parallel Processing?
Such applications are characterized by a very
large amount of numerical computation and/or
a high quantity of input data
In order to deliver sufficient performance for
such applications, we can have manysuch applications, we can have many
processors in a single computer
PP has the potential of being more reliable: if
one processor fails, the system continues to
work at a slightly lower performance
11. Solution is……
The perfect solution is PARALLELISM
In hardware as well as software
12. How parallel processing divide
workload
Parallel computing is an evolution of serial
computing
where the jobs are broken into discrete
parts that can be executed concurrentlyparts that can be executed concurrently
Each part is further broken down to a
series of instructions
Instructions from each part execute
simultaneously on different CPUs
13. Classification Parallel Processor
Architectures
Flynn has classified the computer systems
based on parallelism in the instructions and in
the data streams
These are: These are:
Single instruction, single dataSingle instruction, single data (SISD)(SISD)
Single instruction, multiple dataSingle instruction, multiple data (SIMD(SIMD)
Multiple instruction, single dataMultiple instruction, single data (MISD)(MISD)
Multiple instruction, multiple dataMultiple instruction, multiple data (MIMD)(MIMD)
15. Single instruction, single data (SISD)
A processor that can only do one job at a time
from start to finish.
16. Single instruction, multiple data (SIMD)
A single machine instruction stream
Simultaneous execution on different sets of
data
A large number of processing elements A large number of processing elements
Array and vector processors are the most
common SIMD machines
20. Multiple instruction, single data
(MISD)
In computing, MISD (multiple
instruction, single data)
is a type of parallel computing architecture
where many functional units perform differentwhere many functional units perform different
operations on the same data
21. Multiple instruction, single data
(MISD)
Few actual examples of this class of parallel
computer have ever existed i.e. multiple
frequency filters operating on a single signal
streamstream
24. Multiple instruction, multiple data (MIMD)
Multiple Instruction: every processor may be executing a
different instruction stream
Multiple Data: every processor may be working with a
different data streamdifferent data stream
Distributed systems are MIMD architectures
Most modern computer fall in this category
Examples are: Super Computer, Network Parallel
Computer etc.
27. Advantages of Parallel processing
SaveSave timetime andand costcost
SolveSolve Larger ProblemsLarger Problems (You are ask to do(You are ask to do
1000 Calculus1000 Calculus questions in 1 hour. In facts, I canquestions in 1 hour. In facts, I can
only do 3 question in 1 hours)only do 3 question in 1 hours)only do 3 question in 1 hours)only do 3 question in 1 hours)
ConcurrencyConcurrency (do multiple things at the same(do multiple things at the same
time)time)
Can be made highly faultCan be made highly fault--toleranttolerant
28. Advantages of Parallel processing
Taking advantage of non-local resources
Overcoming memory constraints
Parallel nature of the problem, so parallel
models fit it best
30. Application of Parallel Processing
System
Business Intelligence
Banking, Finance,
Insurance, Risk Analysis
Regression tests for large software Regression tests for large software