Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
GPU v.s. CPU solvers - Roadmap of Comsol for future solvers?
Posted Mar 9, 2011, 4:22 a.m. EST 3 Replies
Please login with a confirmed email address before reporting spam
If you just search the internet for CPU vs GPU, you can get some basics what does it means.
Graphics processing units (GPUs) have, for many years, powered the display of images and motion on computer displays. GPUs are now powerful enough to do more than just move images across the screen. They are capable of performing high-end computations that are the staple of many engineering activities.
Benchmarks that focus on floating point arithmetic, those most often used in these engineering computations, show that GPUs can perform such computations much faster than the traditional central processing units (CPUs) used in today’s workstations—sometimes as much as 20 times faster, depending on the computation
I am just woundering if comsol is planning to include some GPU solver support in future releases?
Graphics processing units (GPUs) have, for many years, powered the display of images and motion on computer displays. GPUs are now powerful enough to do more than just move images across the screen. They are capable of performing high-end computations that are the staple of many engineering activities.
Benchmarks that focus on floating point arithmetic, those most often used in these engineering computations, show that GPUs can perform such computations much faster than the traditional central processing units (CPUs) used in today’s workstations—sometimes as much as 20 times faster, depending on the computation
I am just woundering if comsol is planning to include some GPU solver support in future releases?
3 Replies Last Post Aug 7, 2015, 7:42 a.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi,
there are serveral related threads in this forum.
To my knowledge, CUDA is still not applicable enough, due to the fact that it requires an undeniable big effort for code porting. Besides that, there are contradicting information about the performance improvements (e.g. GPU designed for for single precision operations !?)
However, buying a better WS is always performance improving :)
there are serveral related threads in this forum.
To my knowledge, CUDA is still not applicable enough, due to the fact that it requires an undeniable big effort for code porting. Besides that, there are contradicting information about the performance improvements (e.g. GPU designed for for single precision operations !?)
However, buying a better WS is always performance improving :)
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi,
there are serveral related threads in this forum.
To my knowledge, CUDA is still not applicable enough, due to the fact that it requires an undeniable big effort for code porting. Besides that, there are contradicting information about the performance improvements (e.g. GPU designed for for single precision operations !?)
However, buying a better WS is always performance improving :)
What I can say more than is that CUDA is already double precision ready for new GPUs.
Also, look at Jacket v1.7. They have already ported main solvers to GPU. I don't see why Jacket cannot be used in between comsol calls for solving and assembly, since specially it is for matlab and has an API.
Danial
--
Comsol 4.1
Ubuntu 10.04.1
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
I have recently upgraded my workstation to a double Xeon CPU (E5-2697v2) with 256GB RAM.
I am also planning to install a high end GPU (GTX Titan X with 12GB GDDR5).
And curiously waiting comsol to support GPU :)
I am also planning to install a high end GPU (GTX Titan X with 12GB GDDR5).
And curiously waiting comsol to support GPU :)