<1> Time analysis:

<2> Big-O notation: the order of algorithm.

1. an algorithm is said th be O(N^p)
   if its operation count Q satifies: where C is a constant.

   if p=1 the algorithm is called linear.
   if p=2 the algorithm is called quadratic.

2. An algotithm s called logarithnic O(log n)
   if its operation count Q satifies:
   ex.

   count	2n	4n
   n	2n	4n
   n2	4n2	16n2
   log n	log n + log 2	log n + log16

   Remark:

   1. The order of an algorithm generally is more important than the speed of the processor.
      ex.

      n	O(n) / 3n	O(n2) / n2+2n	O(log n) / log n + 1
      10	30	120	2
      1000	300	10200	3
      10000	3000	1002000	4

   2. A loop that does a fixed amount of operations n times requires O(n) time.
      eg.

      for ( i = 0 ; i < n ; i++ )
          j = j + 1 ;
      

   3. worst-case, average-case, best-case analysis.

      	worst case
      Count the number of operations in	average case
      	bestt case

      ---