The need of parallel computation, Computer Engineering

The Need Of Parallel Computation

With the advancement of computer science (CS), computational speed of the processors has also improved many a time. Moreover, there is certain limitation as we go upwards and face big complex problems. So we have to ready to look for alternatives. The answer denigration in parallel computing. There are two major reasons for using parallel computing: accumulate time and solve larger problems. It is clear that with the increase in amount of processors working in parallel, computation time is clear to reduce. Also, they're some scientific trouble that evens the best processor to takes months or even years to resolve. Moreover, with the application of parallelcomputing these troubles may be resolve in a few hours. Some Other reasons to accept parallel computing are:

i) Cost savings: We can use several cheap computing resources in its place of paying.

ii) It is heavily for a supercomputer.

iii) Overcoming memory constraints: one computer have very finite memory resources. For big problems, using the recollections of multiple computers may conquer this obstacle. So if we join the memory resources of multiple computers then we can simply fulfil the memory requirements of the mass problems.

iv) Limits to serial computing: Physical and practical factors both are pose significant constraints to only building ever quicker serial computers. The speed of a serial computer is totally dependent upon how quick data can travel through hardware. Total limits are the pace of light (3*108 m/sec) and the transmission boundary of copper wire (9*108 m/sec). Rising speeds necessitate increasing proximity of dealing out elements. Secondly, processor technology is allocating an increasing number of transistors to be located on a chip. However, even with atomic-level or molecular components, a limit will be reached on how little components can be made. It is increasingly costly to make a single processor quicker. Using a big number of moderately fast commodity processors to attain the same (or better) presentation is less expensive.

 

Posted Date: 3/1/2013 4:31:19 AM | Location : United States







Related Discussions:- The need of parallel computation, Assignment Help, Ask Question on The need of parallel computation, Get Answer, Expert's Help, The need of parallel computation Discussions

Write discussion on The need of parallel computation
Your posts are moderated
Related Questions
Explain how a critical section avoids Race condition. To prevent Race Condition, concurrent processes should be synchronized. Data consistency needs that only one process m

Example Multi-layer ANN with Sigmoid Units - Artificial intelligence: We will discuss ourselves here with ANNs containing only 1 hidden layer, as this makes describing the back

I n a time division space switch the size of the control memory is N and its Width:  (A) Log 10 M  (B) Log e M  (C) Log N M  (D) Log 2 M Where N are the ou

Build the circuit using the Asynchronous Counter Technique with JK FF and relevantgates capable of executing the counting sequence as {0, 1, 2, 3, 4, 5, 6, 7, 8, 9}Simulate the cir

Explain the uses of thumbwheels Two thumbwheels are usually required to control the screen cursor in its horizontal and vertical position respectively. As the name implies,

Q. Explain Physical Characteristics of magnetic disk? Figure below lists main features that differentiate among different types of magnetic disks. First head may either be fixe

Byteland county is very famous for luminous jewels. Luminous jewels are used in making beautiful necklaces. A necklace consists of various luminous jewels of particular colour. Nec

What are the types of container classes in C++?  Ans) A container class is a class that is used to hold objects in memory/external storage. A container class behaves as a ge

What are primary keys and foreign keys? Primary keys are the unique identifiers for every row. They must have unique values and cannot be null. Due to their significance in rel

Why a function can n ot call a task As functions doesn't consume time,it can do any operation which doesn't consume time. Mostly tasks are written that consumes time. So a ta