A sorting method with “Big-oh” complexity 0(n logn) spends exactly 1 millisecond to sort 1000 data items.assuming that time T(n) of sorting n items is directly proportional to nlogn ,that is T(n)=cnlogn, derive a formula for T(n),given the time T(n) for sorting N items and estimate how long this method will sort 1,000,000 items.

# Sorting method with Big-oh complexity

**Anmol_Binani**#2

T(n) = c *n logn

=> 1 msec = c *1000 log(1000)

=> c = 1 / (1,000,000 * log1000)

Now,

T(n) = c * (1,000,000) log(1,000,000)

substitute the value of ‘c’, we get:

**T(n) = 2 milli sec**