1) velocity of the machine strolling this system
2) language in which this system changed into written. As an instance, packages written in meeting language usually run faster than those written in c or c++, which in flip tend to run faster than the ones written in java.
3) efficiency of the compiler that created the program
4) the dimensions of the input: processing 1000 records will take extra time than processing 10 statistics.
5) organisation of the input: if the item we are searching for is at the pinnacle of the list, it’s going to take less time to discover it than if it’s far at the bottom.
The first three items inside the list are problematic. We don’t want to apply an genuine dimension of strolling time: to say that a particular set of rules written in java and strolling on a pentium iv takes a few wide variety of milliseconds to run tells us not anything approximately the general time performance of the algorithm, because the dimension is particular to a given environment. The measurement might be of no need to someone in a exclusive environment. We need a preferred metric for the time performance of an algorithm; one this is impartial of processor or language speeds, or compiler efficiency.