Assuming you are familiar with scientific readings, I'd recommend to read papers on algorithms and data structures on a regular basis. There are tons of it and they normally discuss complexity and shortcomings as well as comparisons to other algorithms. Personally I find it much harder, to understand algorithms, other people wrote, than coming up with my own algorithms. I like to print such papers (normally 4-12 pages) and carry them with me (only one at a time), wherever I go, so I can read a few lines, whenever I have to wait somewhere.