2. Contents
1- INTRODUCTION
2- Numeric Weather Prediction(NWP)
3- Oceanography
4-Astrophysics
5-Socio Economics
6-Finite Element Analysis(FEA)
7-Artificial Intelligence and Automation (AI)
8-Seismic Exploration
9- Genetic Engineering
10-Weapon Research and Defense
11-Medical Applications
12-Remote Sensing Applications
13-Energy Resource Exploration
14-FPGAs power applications performance
2
15- Conclusion
3. 1- INTRODUCTION
Parallel processing refers to the speeding up a computational task by
dividing it into smaller jobs across multiple processors. Notable
applications for parallel processing (also known as parallel
computing) include computational astrophysics, geo processing (or
seismic surveying), climate modeling, agriculture estimates, financial
risk management, video color correction, computational fluid
dynamics, medical imaging and drug discovery.
3
4. Parallel computing is now being used extensively
around the world, in a wide variety of applications
4
5. 2- Numeric Weather Prediction(NWP)
NWP uses mathematical models of
atmosphere and oceans
Taking current observations of weather
and processing these data with
computer models to forecast the future state
of weather.
Uses data assimilation to produce outputs
5
6. 3- Oceanography
Used to study wealth of ocean using
multiprocessors having large
Computational power with low power
requirements.
ROMS were used originally but now
MPI programming methods are used.
6
7. 4- Astrophysics
Computational astrophysics refers to the
methods and computing tools developed and
used in astrophysics research.
PIC ,PM and n-body simulations are
different important techniques for
computational astrophysics.
7
8. 5- Socio Economics
Parallel processing is used for modelling of a economy of a
nation/world.
Programs system which involves cluster computing device to
implement parallel algorithms of scenario calculations ,optimization
are used in such economic models.
Such program system serves for conducting multi-scenario
calculations to design a suitable development strategy for a region.
8
9. 6- Finite Element Analysis(FEA)
FEA is a numeric method commonly
used for multiphasic problem.
Used in design of huge structures like
ships, dams, supersonic jets etc.
In FEA extremely large amount of partial
differential equations are to solved
concurrently and hence parallel processing
elements are used.
9
10. 7- Artificial Intelligence and Automation
(AI)
AI is the intelligence exhibited by machines or software.
AI systems requires large amount of parallel computing for which
they are used.
Four types
1.Image processing
2.Expert Systems
3.Natural Language Processing(NLP)
4.Pattern Recognition
10
11. 8- Seismic Exploration
It is a method of exploration that uses the
principles of seismology to estimate the properties
of the Earth's subsurface from reflected seismic
waves.
When a seismic wave travelling through the
Earth encounters an interface between two
materials, some of the wave energy will reflect off
the interface which are analyzed to determine
underground strata.
Such large number of reflected waves are
analyzed using sensors and parallel architecture
11
12. 9- Genetic Engineering
It is the direct manipulation of an
organism's genome using biotechnology
for eg. Dna sequence analysis.
Several of these analysis produce huge
amounts of information which becomes
difficult to handle using single
processing units because of which
parallel processing algorithms are used
12
13. 10- Weapon Research and Defense
Computer clusters are used in simulations that show a nuclear
weapon's performance in precise molecular detail, tools that are used
for national defense.
Parallel computing is required which are needed to more efficiently
certify nuclear weapons, to accurately show molecular-scale reactions
taking place over milliseconds, or thousandths of a second.
They are also used in plutonium research to study its behavior under
high pressure whose alloys are then used for making explosives
13
14. 11- Medical Applications
Parallel computing is used in medical
image processing
Used for scanning human body and
scanning human brain
Used in MRI reconstruction
Used for vertebra detection and
segmentation in X-ray images
Used for brain fiber tracking
14
15. 12- Remote Sensing Applications
It is a software application that processes remote
sensing data.
Remote sensing applications read specialized file
formats that contain sensor image data,
georeferencing information, and sensor metadata.
Computer analysis of such remotely sensed earth
resources data has many applications in agriculture,
forestry etc.
Explosive amounts of pictorial information needs
to be processed in this area.
15
16. 13- Energy Resource Exploration
Resource Exploration is a method to
gather and manage information about
energy resources like oil, natural gas etc.
Computers here help in the discovery
and management of such energy
resources.
This sector maintains the records of
global energy crisis and also helps in
ensuring nuclear reactor safety
16
17. 14- FPGAs power applications performance
FPGAs offer tremendous performance potential. They can
support in number of different parallel computation applications
and implemented in single clock execution time. If FPGAs are
reprogrammable then they can provide on chip facility for a
number of applications. Due to the presence of on-chip memory
facilitate coprocessor logic’s memory access bandwidth is not
restricted to the number of I/O pins present in the devices.
17
18. FPGA application design techniques
1. Use an algorithm optimal for FPGAs
2. Use a computing mode appropriate for FPGAs
3. Use appropriate FPGA structures
4. Living with Amdahl’s law
5. Hide latency of independent functions
6. Use rate techniques to remove bottlenecks
18
22. Conclusion
1- Compared with serial computing, parallel computing is
more suitable for modeling, simulation and understanding of
complex real-world phenomena.
2- Save time and / or money.
3- Solve additional / large complex problems Example: Web
search engines / databases handle millions of transactions
every second.
22