Speeding up the calculations in general

[fumitoh]

(Continued from #45)
Models based on BasicTerm_S or BasicTerm_SE are slow. There are couple of ways to improve the speed.

1. Profile the model using get_stacktrace / stop_stacktrace and find formulas that are taking time and rewrite them to run faster. This page explains how to use the functions.

2. Develop the models from BasicTerm_M or BasicTerm_ME instead. These models are designed to run fast. On the other hand, BasicTerm_S or BasicTerm_SE are more for model validations.

3. Run the model in parallel using ipyparallel. Split a run in N runs by splitting the in-force file, where N is the number of logical CPU cores on your machine. This blog post discusses how to do it in more detail. If there were no overhead, the run should get N times faster, but in reallity, the improvment should be lower than N.