THE NEED OF PARALLEL COMPUTATION

 With the growth of computer science, computational pace of the processors has also increased many a times. Though, there are definite constraints as we go upwards and face large complicated problems. So we would have to look for alternatives. The answer lies in the parallel computing. There are two main reasons for using parallel computing: solve larger problems and save time. It is clear that with the raise in some processors working in parallel, computation time is bound to decrease. In addition, they're some scientific problems which even the fastest processor to takes months or even years to solve. On the other hand, with the application of parallel computing these problems can be solved in a few hours. Other reasons to approve parallel computing are: 

i)  Cost savings: We can make use of multiple cheap computing resources instead of paying heavily for a supercomputer. 

 ii) Overcoming memory constraints: Single computers have exceptionally finite memory resources. For big problems, using the memories of multiple computers may overcome this obstacle. So if we merge the memory resources of multiple computers then we may effortlessly fulfill the memory requirements of the large size problems.

iii) Limits to serial computing: Both practical and physical factors pose important constraints to basically building ever quicker serial computers. The pace of a serial computer is straightforwardly dependent upon how fast data can move through hardware. Absolute limits are speed of light (3*108 m/sec) plus transmission limit of copper wire (9*108 m/sec). Ever-increasing speeds necessitate increasing proximity of processing elements.  Secondly, processor technology is permitting an increasing number of transistors to be positioned on a chip. Though, even with molecular or atomic-level components, a limit would be reached on how small components can be made. It is increasingly costly to create a single processor faster. Using a larger number of moderately rapid commodity processors to get the same (or better) performance is less expensive.