The fundamental difference between problems in discrete mathematics and problems in calculus or algebra is the kind of data the problems deal with. Calculus and algebra deal with information on an infinite continuum, but discrete mathematics deals with "discrete" information, which is independent and finite -- like bits of data are in computers. A simple problem might ask how many 4-digit numbers can be made using only the numbers 1, 2 and 3. Such a problem would be solved by considering the possibilities for each digit individually.
Discrete mathematics deals fundamentally with the development of algorithms. Algorithms are explicit, step-by-step procedures for performing calculations or solving problems. An example procedure for finding the number of possible 4-digit combinations of 1, 2 and 3 would be to first count the possible values for the first digit of a 4-digit number composed of 1's, 2's and 3's, to repeat this for the next three digits, and to multiply the possible values for each digit -- 3 * 3 * 3 * 3 -- for the solution, 891.
Discrete mathematics is concerned with more than just developing algorithms, however. One of the goals of discrete math is to make sure finding solutions doesn't take more work or time than is practical or permissible. For example, finding how many 4-digit numbers can be made using only 1's, 2's and 3's can be done by listing and then counting all possible 4-digit combinations of 1, 2 and 3, but such a solution would take far more work than considering the digits individually and multiplying.
Because discrete mathematics is most used in computer science, its ultimate goal is to allow computers to solve problems. This highlights the importance of algorithms. Because computers can only follow instructions -- with no understanding of what they're doing or why -- those instructions, which are algorithms, must be precise enough to find the solution and simple enough to not use excessive processing power or physical storage.