I get to read quite a few papers in the computer vision and robotics field. Most of those papers are too high level or represent work that I’ll probably never get a chance to do. Recently I ran into 2 papers that have application in my everyday work. I’d like to share them with you.
What every programmer should know about memory
The paper by Ulrich Drepper is available here. This is one of those papers that make me feel embarrassed about how much I didn’t understand. This paper will cover the workings of memory, timing pitfalls, multi-processor architecture, and what programmers can do about this whole mess. I’m still working on this paper as it is quite lengthy but it is totally worth it.
What every programmer should know about floating-point arithmetic
The paper by David Goldberg is available here. The previous paper was modelled after this article. I haven’t started reading this one yet. If this inspired Ulrich, I assume this is of equal or better caliber to the other.