What are the factors that influence the performance of algorithm?

Answer 1

The performance of an algorithm depends primarily on two concepts: concurrency and number of operations per input unit. Concurrency occurs when an algorithm can split its data into parts and work on each part in parallel, such as a binary sort algorithm that uses "threads" to break its work into equal loads. JPEG algorithms can operate on 8x8 blocks of pixels, and are well-suited to concurrent algorithms, while solving a long algebraic equation may not be suitable for concurrent operation if the results of each step determine the following step.

The second factor of performance is known as the Big-O notation. This is summarized as an algorithm's response to the size of the input, where O(1) is fixed time (and fastest), and O(n) is a linear increase to input size, and larger values, such as O(n^2) representing exponential growth of processing time dependent on input size, or O(n!) representing factorial growth (very poor scaling). Other models that represent scaling also exist, but Big-O appears to be one of the most common models for conveying the cost of an algorithm.