COMPUTER SCIENCE AND ENGINEERING
ALGORITHMS
|
Question
[CLICK ON ANY CHOICE TO KNOW THE RIGHT ANSWER]
|
|
|
typically are slower programs than Sequential
|
|
|
cannot be slowed down even if only one of many devices is slow
|
|
|
can improve every single time more processors are added
|
|
|
are often times the fastest option, but there is a limit
|
Detailed explanation-1: -One is due to additional total hardware resources when running in parallel. The other from changing data access patterns in parallel. There are limited amounts of faster memory attached to each processor called cache.
Detailed explanation-2: -Practically, when a program is executed in parallel, the hypothesis that the parallel program will run faster is not always satisfied. If the main goal of parallelizing a serial program is to obtain a faster run then the main criterion to be considered is the speedup gained from parallelization.
Detailed explanation-3: -Applications rarely parallelize perfectly because part of the application is serial. This serial component imposes a limit on the amount of parallelization the application can use. Amdahl’s law describes the speedup you can get from parallel processing.
Detailed explanation-4: -The speedup of a program from parallelization is limited by how much of the program can be parallelized. For example, if 90% of the program can be parallelized, the theoretical maximum speedup using parallel computing would be 10 times no matter how many processors are used.