## Background:

In the finite element numerical simulation, the last step is to calculate the linear equations of the overall stiffness matrix: K×d=F .

In julia, the method of solving linear equations can be directly “`\`

”:

`d = K \ F`

## Question:

When the matrix is very large, what method can be used to speed up the calculation?

There are the following requirements on the method:

- The calculation accuracy is best not to change
- My computer is a 40-core workstation, maybe I can consider multi-core parallelism? (I’m not sure if there is such a way)
- All the methods provided are preferably simple to implement (I don’t need to change too much code )

Welcome everyone to make suggestions, or what information can be used for reference, thank you all.