MATLAB: GpuArray for vpa (variable precision arithmetic)

gpuarrayParallel Computing Toolboxvpa

Problem:

I have big matrices with elements defined through variable precision arithmetic of the symbolic toolbox (vpa(x)).

Extracting the eigenvalues out of these matrices takes about 7 minutes.

operations that imply inversion-ish operations (C=A\B) also take some minutes.

Question:

Could I have any benefit from gpu computation?

To be clearer on what I don't understand:

I know that symbolic maths are not supported by GPU computing but in my case the variable is not really symbolic, it just has higher precision.

Thank you for any clarification

Best Answer

In this case you are going to have to sacrifice something, if you use VPA the matrices become symbolic, therefore all the operations you do with them are going to slow down, GPU does not support these matrices so it would not give you anything.

On the other hand if you stay with "normal" matrices you will have a maximum of 15 digits after the decimal point, and using the GPU could give you something depending on the operations you are doing

If you really need too much precision you can search about Pari / GP that you could integrate with Matlab, or also Accelerating MATLAB Code Using Codegen and Mex Files

Related Solutions

MATLAB: Benchmark performance of GPU computing of Monte-Carlo simulation

Which GPU do you have?

There are overheads to communicating with the GPU, and it could be the case that only your single precision performance is enough to overcome those overheads. Entry-level NVIDIA GPU are better on single precision than double precision because graphics is usually done in single precision and the GPU cards in entry systems are usually there to support graphics rather than to give a platform for high performance computation.

MATLAB: How to use VPA (variable precision arithmetic) in calculating HH band via dwt on an image

You can not do this. dwt2 accepts only double as input and outputs double. vpa works on symbolic variables and returns a symbolic variable:

>> vpa(pi)
 
ans =
 
3.1415926535897932384626433832795
 
>> whos ans
  Name      Size            Bytes  Class    Attributes
  ans       1x1                 8  sym

So you will either get an error if you try to pass a symbolic variable to dwt2 or you will have to cast it to double, which also does not help you. What you can do is show 32 digits of the result calculated as double:

  >> a = pi/2
a =
    1.5708
>> sprintf('%.32g',a)
ans =
    '1.5707963267948965579989817342721'
>> whos
  Name      Size            Bytes  Class     Attributes
  a         1x1                 8  double              
  ans       1x33               66  char

Maybe this helps.

Best Answer

Related Solutions

MATLAB: Benchmark performance of GPU computing of Monte-Carlo simulation

MATLAB: How to use VPA (variable precision arithmetic) in calculating HH band via dwt on an image

Related Question