Parallel and Distributed Computing (College Board AP® Computer Science Principles): Flashcards
0Still learning
Know0
- FrontComputing Models
Define sequential computing.
Was this flashcard helpful?
Cards in this collection (22)
Define sequential computing.
Sequential computing is a computational model in which operations are performed in order, one at a time.
Define parallel computing.
Parallel computing is a computational model where the program is broken into multiple smaller sequential operations, some of which are performed simultaneously.
Define distributed computing.
Distributed computing is a computational model in which multiple devices are used to run a program.
True or False?
Distributed computing means using several processors inside a single computer.
False.
Using several processors in one computer is parallel computing. Distributed computing uses multiple devices.
How can the efficiency of two computing solutions be compared?
Compare the time it takes each solution to perform the same task.
A sequential solution takes as long as the ___ of all of its steps.
A sequential solution takes as long as the sum of all of its steps.
How long does a parallel solution take?
A parallel solution takes as long as its sequential tasks plus the longest of its parallel tasks.
Four tasks take 3, 5, 2, and 4 seconds. How long do they take when run sequentially?
Run sequentially, the tasks take 14 seconds (3 + 5 + 2 + 4).
Processor 1 runs tasks of 3 and 5 seconds. Processor 2 runs tasks of 2 and 4 seconds at the same time. How long does the parallel solution take?
The parallel solution takes 8 seconds, the time of the longest processor (3 + 5).
A solution has a 10-second sequential portion, followed by tasks of 20 and 30 seconds running in parallel on two processors. How long does it take?
The solution takes 40 seconds: the 10-second sequential portion plus the longest parallel task (30 seconds).
Define speedup.
The speedup of a parallel solution is the sequential time divided by the parallel time.
A task takes 20 seconds sequentially and 8 seconds in parallel. What is the speedup?
The speedup is 2.5 (20 / 8).
A speedup of __ would mean no improvement over the sequential approach.
A speedup of 1 would mean no improvement over the sequential approach.
What two portions does a parallel computing solution consist of?
A parallel computing solution consists of a parallel portion and a sequential portion.
How do parallel solutions compare with sequential solutions when scaling?
Parallel solutions scale more effectively than sequential solutions.
Name two reasons a problem might be impossible to solve on a single computer.
A single computer may not cope because of the processing time or storage needs involved.
Apart from solving problems a single computer cannot, what other benefit does distributed computing give?
Distributed computing allows much larger problems to be solved quicker than on a single computer.
Parallel computing completes tasks faster by performing operations at the ____ time on multiple processors.
Parallel computing completes tasks faster by performing operations at the same time on multiple processors.
What limits the efficiency of a solution as more parallel computing is added?
Efficiency is still limited by the sequential portion of the solution.
Name two tasks that typically must run sequentially in a parallel solution.
Tasks that must run sequentially include initializing data before processing and combining results after processing.
The sequential portion sets a _______ time that the solution cannot go below, regardless of how many processors are added.
The sequential portion sets a minimum time that the solution cannot go below, regardless of how many processors are added.
A parallel solution includes a 25-second sequential portion. What is the lower limit on its run time, however many processors are added?
The solution can never take less than 25 seconds, because the sequential portion cannot be parallelized.
Sign up to unlock flashcards
or