CPUs and architecture

If 90% of a program can be parallelized, what is the theoretical maximum speedup when it runs on an unlimited number of processors?

  1. A10 times
  2. B90 times
  3. CInfinite
Show answer and explanation

Answer: A. 10 times

According to Amdahl's law, the part that cannot be parallelized (10%) does not get shorter no matter how many processors are added, so the total execution time cannot fall below 10% of the original. The maximum speedup is therefore 1 ÷ 0.1 = 10 times. This shows the importance of reducing the non-parallelizable part.

Practice this topic Back to topic

More CPUs and architecture questions

  1. Which scheme performs the same operation on multiple data items simultaneously with a single instruction?
  2. Which CPU configuration combines high-performance cores and power-efficient cores and uses them according to the workload?
  3. Which technique examines instruction dependencies and executes ready instructions in an order different from the program order to improve performance?
  4. Which CPU mechanism temporarily suspends the current task to give priority to handling requests such as those from external devices?
  5. Which mechanism transfers data directly between peripherals and memory without going through the CPU?