Re: Matlab Vectorisation Speed - How is it done in c++?

On 17 Des, 01:05, Phil Winder <philipwin...@xxxxxxxxxxxxxx> wrote:

Im currently porting some matlab algorithms to c++ code. The test
code I have is testing the vector math capabilities and how fast they
can go. I have found that it can be very, very fast and I am
strugglling to reproduce the speed in c++. How does matlab do it? And
how can it be reproduced in c++?

Beating the performance of vectorized Matlab code is very hard, and
usually not worth the effort.

Matlab makes calls to optimized C and Fortran libraries such as blas/
atlas, lapack and fftw. You cannot duplicate their efficacies in C++
for at least two reasons:

1. There are issues related to the language syntax that makes Fortran
particularly easy to optimize for compilers, such as lack of pointer
aliasing. This is particularly important for optimal allocation of
registers when the CPU goes into a tight loop.

2. A lot of effort have been put into making these libraries as fast
as possible. This includes optimal use of cache and branch prediction.
Duplicating these efforts on your own is going to take the rest of
your life to complete.

My advice would be this:

If you want speed in your C++ app, link and call the same libraries as
Matlab do. Most of them are available for free. Give the C++ compiler
pointer aliasing hints wherever possible.

In addition:

Use optimization level 3 on numerical code and level 2 on non-
numerical code. Process the data in chunks that fit in your L1 cache.
Force the CPU to prefetch if you know it will help. Page-align your
arrays in RAM. Memory access is terribly slow, traverse as few times
as possible. Never use strided memory access. Manually unroll tight
loops. Exploit arithmetic pipelining of four subsequent operations.
Avoid divisions, transform to a multiplcation. Exploit multiple CPUs:
use MPI or OpenMP, forkjoin with labour-sharing threadpools, etc. Use
inline assembly to access SIMD parallel registers.

Also remember Hoare's statement about optimization, quoted by D.
Knuth: "Premature optimization is the root of all evil in computer
programming." Profile your code. Direct your optimizations to the
important bottlenecks. They are likely to be few. Do as much as you
can with the bottlenecks, and never mind the reminding 90% of your
code.

But before you begin: ask yourself if the hard work is goint to be
worth the effort.

.

Relevant Pages

  • Re: Matlab Vectorisation Speed - How is it done in c++?
    ... code I have is testing the vector math capabilities and how fast they ... How does matlab do it? ... Memory access is terribly slow, ... Also remember Hoare's statement about optimization, ...
    (comp.soft-sys.matlab)
  • Re: Cost function for a model instead of a function?
    ... I've been using MATLAB for about one year, but do not have true knowledge ... after simulation, calculates measurements such as mean distance, maximum ... are then passed through a T-Test to determine statistical likeness. ... The Optimization Toolbox solvers do not require you to be able to write down ...
    (comp.soft-sys.matlab)
  • Re: Matlab Vectorisation Speed - How is it done in c++?
    ... Beating the performance of vectorized Matlab code is very ... Matlab makes calls to optimized C and Fortran libraries ... Use optimization level 3 on numerical code and level 2 on non- ...
    (comp.soft-sys.matlab)
  • Re: can I use matlab for this optimization problem?
    ... Optimization Toolbox to solve the problem you describe and then call ... your optimization routine from your C++ application using the MATLAB ... For developing your optimization routine, a good place to start is to ...
    (comp.soft-sys.matlab)
  • Re: Optimal programming advice
    ... When you do get to working on optimization, the first thing to optimize is the algorithm instead of the code. ... In todays architectures, memory access is often a dominating factor in speed, often outweighing the actual "computations". ... This is one of those things that have changed over time, so you'll often get counterproductive advice if you look at things from 20 years ago. ...
    (comp.lang.fortran)