speed up code using servers

6 visualizaciones (últimos 30 días)
Kim
Kim el 8 de Sept. de 2014
Comentada: Kim el 26 de Oct. de 2014
I have a code that takes more than a day to run. It has a lot of nested loops so I cant use parfor to speed it up much. Can I hire computing time on a superfast computer and run it there? I dont want to use the parallel computing toolbox, because it wont help me much. I just need a very fast computer (better than my desktop i7 processor). Would it help if I use Amazons server or is that only useful if you can benefit from parallel computing? Thank you!
  2 comentarios
Iain
Iain el 8 de Sept. de 2014
Matlab uses a single core. Opening up all of your cores (An i7 has 8?) ought to speed up your execution time by about a factor of 6-7 (if you use 8 cores) when you use parfor. - If you can use parfor that is.
You might find improvements by using graphical processors. "gpuArray" is a function you'll find in the help documentation.
Kim
Kim el 26 de Oct. de 2014
Sorry for the late reply, it took me a while to look into all of this. For me using parfor is not going to help much I think, as I have many forloops 'over' each other. I found the following server though which should speed up things even without using parfor: https://www.cac.cornell.edu/redcloud/

Iniciar sesión para comentar.

Respuestas (0)

Categorías

Más información sobre Loops and Conditional Statements en Help Center y File Exchange.

Etiquetas

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by