When timing a critical function for an algorithm, using platform time functions such as get time of day (), the result is unpredictable; there is significant variance in the measured time between different runs of the benchmark. Which of the following strategies would improve the accuracy of the measurement?
Submit