A training system using 8 processing devices achieves a throughput of 1,000 samples per second. When the team scales up to 32 processing devices, the throughput becomes 3,200 samples per second. Based on this information, did the system achieve perfect linear scalability? Explain your reasoning.

Google

The primary objective of parallel processing in distributed training is to achieve linear scalability. This means that the system's efficiency, measured by the number of samples processed per unit of time, should increase in direct proportion to the number of processing devices used.

Goal of Parallel Processing: Linear Scalability

A team is evaluating the performance of their distributed training setup. They measure the number of data samples processed per second as they increase the number of processing devices. Which of the following outcomes best illustrates the ideal goal of linear scalability?

A research lab is testing two different distributed training configurations, System A and System B, to find the more efficient setup. Based on the performance data below, which system better achieves the ideal goal of parallel processing? Justify your answer by explaining how the data for each system relates to this goal.

Learn Before

Related