|You can find the latest edition of the course notes
here. Version history:
The picture illustrates approximate limits on where each category of time complexity can result in solutions worth waiting for; the area hints at how many useful algorithms is likely to exist within that complexity class. For example: O(kn) algorithms – bottom left – are only useful for very small n and there are very few that are useful for non-trivial problems, especially with the large data sets in bioinformatics. Algorithms that would be too slow for a particular scale of data may lead us off the picture to approximation methods.