Question

In a C++ CPU-bound simulation that I’m writing, I’ve traced a bottleneck via valgrind in my program to cmath::exp. It currently eats over 40% of my simulation time. I can bound the input to a relatively small domain, but I’d like to control the accuracy. I was considering moving to a LUT (lookup table) to replace exp but I’m not quite sure how to do this the “right way”(tm). Concerns I have:

1. Large lookup tables will not fit in the cache thus slowing access
2. Best way to convert a double input to an integer for access into the lookup table
3. Does the answer to (2) depend on the slope of the input function?
4. Am I reinventing the wheel – has this already been done before?

What is the best way to implement/(include from a library) a LUT for exp?

Answer 1

A very similar question has been asked before. Here’s my answer:

• Fast Exp calculation: possible to improve accuracy without losing too much performance?

The approach was suggested by the author of that question, and I was able to implement it efficiently (small lookup table and minimal extra work after lookup). It’s in C#, but translation to C++ should be straightforward and I can help if you run into trouble.