I'm trying to install caffe and I wonder if I can use cuDNN with AMD/OpenCL.
Because my graphic card is AMD
https://github.com/BVLC/caffe/tree/opencl
I'm afraid this won't work: cudnn is an extension of cuda, which is a propriety of NVIDIA. Thus, a non-NVIDIA GPU does not support CUDA and thus does not support cuDNN.
With a non NVIDIA card, you cannot run CUDA code (main caffe branch), but you should be able to enjoy opencl GPU code. You should give the opencl branch a chance.
The short answer is that if your graphics card is AMD then you'll have to use OpenCL, not cuDNN. You cannot make them work together.
cuDNN and OpenCL are competition, and so it doesn't even make sense to try and use them together.
If instead you are asking if you can use NVIDIA's cuDNN library on AMD hardware, the answer is no. It just isn't compatible. cuDNN was made specifically to work on the NVIDIA hardware and take advantage of the unique properties of that chip set.
There is an OpenCL variant of cuDNN from Intel:
https://github.com/01org/clDNN/
Since it is OpenCL-based, should work also on AMD GPUs (although I haven't tested it myself)
I am afraid if you can really use it for AMD graphics card since clDNN is built for DL inference in particular for Intel graphics cards (HD and Iris). Check e.g., OpenVINO toolkit (designed by Intel), it uses clDNN under the hood for GPU plugins to accelerate inference on the Intel GPUs.
From the GPU plugin page, it says:
clDNN is an open source performance library for Deep Learning (DL) applications intended for acceleration of Deep Learning Inference on Intel® Processor Graphics including Intel® HD Graphics and Intel® Iris® Graphics.
Related
I currently do not have an Nvidia GPU on my laptop but I still want to be able to speed up training. Per the Flux docs, it says Nvidia GPU's are supported out of the box but doesn't mention AMD GPU's at all. Is it possible to work with Flux on an AMD GPU?
This should in principle be possible, since AMDGPU.jl provides a similar interface for AMD GPUs as CUDA.jl does for NVIDIA GPUs, and Flux is claimed to be agnostic to array types. However, at best this will only work on Linux, since AMDGPU.jl relies on AMD's ROCm platform, and ROCm is only supported on Linux, since ROCm is indeed specifically built around the Linux kernel.
What is the best way to perform fft with opencl? (my gpu is intel HD graphics 4600)
I found clfft library, but stucked on installing it. I read documentation, but don't understand couple things. It requires some dependencies, one of them is AMD APP SDK, and as I can guess its only for amd gpu's? Can I use it with intel hd graphics?
I have seen that AMD APP SDK samples work on a machine having only Intel CPU.
How can this happen? How does the compiler target a different machine architecture?
Do I not need Intel's set of compilers for running the code on the intel CPU?
I think if we have to run an OpenCL application on a specific hardware, I have to (re)compile it using device's vendor specifics compiler.
Where is my understanding wrong?
Firstly, OpenCL is built to work on CPU's and GPU's. You can compile and run the same source code on either type of device. However, its very likely that CPU code will be sub-optimal for a GPU and vice-versa.
AMD H/W is 7% - 14% of total x86/x64 CPU's. So AMD must develop compilers for both AMD and Intel chips to be relevant. AMD have history developing compilers for both sets of chips. Conversely, Intel have developed compilers that either don't work on AMD chips or don't work that well. That's no surprise.
With OpenCL, the AMD APP SDK is the most flexible it will work well on AMD and Intel CPU's and AMD GPUs. Intel's OpenCL SDK doesn't even install on AMD x86 H/W.
If you compile an OpenCL program to binary, you can save and reuse it as long as it matches the OpenCL Platform and Device that created it. So, if you compile for one device and use on another you are very likely to get an error.
The power of OpenCL is abstracting the underlaying hardware and offer massive, parallel and heterogeneous computing power.
Some SDKs and platforms offers some specific features to "optimize" the code, i honestly think that such features are just marketing and they introduce boilerplate code making the application less portable.
There are also some pseudo-new technologies that are just wrappers to OpenCL or they are really similar in the concept like the Intel quick sync.
About Intel i should say that at the first place they were supporting all the iCore generation and even some C2D, now the new SDK only support the 3rd iCore generation, i don't get their strategy honestly, probably Intel is the last option if you want to adopt OpenCL and targeting the biggest possible audience, also their SDK doesn't seems to be really good at all .
Stick with the standard and you will avoid both possible legal and performance issues and your code will also be more portable.
The bottom line is that the AMD SDK includes a compiler for targeting x86 CPUs for OpenCL. That means that even though you are running an Intel CPU the generated code will run on it. It's the same concept as compiling a C program to run on an x86 CPU: it works on Intel and AMD CPUs (or any that implement the x86 instruction set).
The vendor's compiler might have specific optimizations, like user827992 mentions, but in my experience the performance of AMD's CPU compiler isn't that bad when running on an Intel CPU. I haven't tried Intel's OpenCL implementation.
It is true that for some (maybe most in the future) hardware, only the vendor's compiler will support it. AMD's SDK won't build code that will run on an NVIDIA card, and vice-versa. CPUs happen to be a bit of a special case in that the basic instruction set is so widely deployed that the CPU compiler will work on most machines you're likely to come in contact with.
With some friends we want to use openCL. For this we look to buy a new computer, but we asked us the best between AMD and Intel for use of openCL. The graphics card will be a Nvidia and we don't have choice on the graphic card, so we start to want buy an intel cpu, but after some research we figure out that may be AMD cpu are better with openCL. We didn't find benchmarks which compare the both.
So here is our questions:
Is AMD better than Intel with openCL?
Is it a matter to have a Nvidia card with an AMD cpu for the performance of openCL?
Thank you,
GrWEn
You shouldn't care as much about what CPU you use as much as what GPU you use. You would need to choose between an AMD/ATI GPU or nVidia GPU.
I would personally recommend an nVidia GPU as, in addition to OpenCL support, you can experiment with their more proprietary CUDA technology which offers a far richer development experience than OpenCL does today. While you're at it take a look at the new AMP technology that was just announced by Microsoft for C++ which aims to bring language extensions akin to nVidia's CUDA. nVidia also has offerings for the enterprise with their Tesla GPUs with several vendors offering GPU clusters and you can even get a GPU compute cluster on Amazon EC2 now which is all based on nVidia hardware.
You want to buy a new computer with your friends? What kind of project do you plan to do? The question about the hardware is answered with the needs you have. If you give some more information, we can provide better suggestions.
As written before, the CPU is not the important point as long as you do not want to buy a multiprocessor multicore system like 4 Quadprocessors. The difference in performance is mostly the differences of the GPUs used and there you can find different cards for all needs. From a cheap GPU to the nVidia Tesla cards.
It is definitely not a problem to run a nVidia board on a AMD system. I do it here. You also can use the OpenCL devices from the AMD Multicore CPU and the nVidia GPU in parallel.
You should pay attention: If you plan to buy a potent system to run your software (like a webserver), every developer of OpenCL software needs a system for testing. So every developer needs at least a modern multi-core CPU with an OpenCL SDK. Where the OpenCL kernels are developed does not matter. OpenCL is platform independed.
Both Intel and AMD have good OpenCL-support for their CPUs, so currently it does not really matter which you cooose. If you want to use the embedded GPU on AMD Fusion or Intel SandyBridge, then I suggest you go for Fusion since Intel does not have a driver for their GPUs (yet). Depending on what you are going to use OpenCL for, I could suggest a GPU - sometimes NVidia is faster, sometimes AMD.
AMP, CUDA, RenderScript and the many, many others all work nice but they don't work on all hardware as OpenCL does. CUDA certainly has advantages, but in the time you have learnt openCL I can assure you the tools around OpenCL have catched up.
The CPU has no influence on GPU OpenCL performance.
You might also want to try running the OpenCL kernels on CPU. Checkout the Intel OpenCL compiler beta. You can even run kernels on both CPU and GPU.
I googled this topic and didn't find anything new. I am aware of Nvidia's FFT implementation which is great, but for CUDA only. AMD just released their implementation, but it doesn't work on Nvidia cards. Apple has an older and slower implementation. Are there any other good FFT libraries out there? It would be nice if there was an implementation that was meant to work on Nvidia and AMD cards and other possible platforms and is being actively maintained.
The AMD clAmdFft library should work on NVidia GPUs.
I was involved in the development and I know that was the intention. The code was written to the OpenCL standard and doesn't use any proprietary tricks. Of course, AMD didn't do QA testing on NVidia hardware. It could be that NVidia's OpenCL implementation isn't quite 100% compliant to the standard yet. Or it could be something about your particular video card.
Please post more details here as to exacly what is happening. You should also post that information in the AMD developer forums as a bug.
AMD recently released an OpenCL SDK for their CPUs as wel as GPUs. Included in it are FFT and BLAS libraries. You can go to the bottom of the page to find out about the supported devices.
But I am not really sure about the performance.
Not yet - but there is a project to port the GSL (Gnu Scientific Library) to opencl
http://gsl-cl.sourceforge.net/
I know Apple has released an OpenCL FFT package, but I don't know much about it. I've heard that they make the source available.