Ralph support OpenCL ?

Message boards : Number crunching : Ralph support OpenCL ?

To post messages, you must log in.

Previous · 1 . . . 4 · 5 · 6 · 7 · 8 · Next

AuthorMessage
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7178 - Posted: 30 Mar 2022, 13:08:36 UTC

ID: 7178 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7230 - Posted: 25 Jun 2022, 8:22:28 UTC

Pocl 3.0 released with initial Opencl 3.0 support (and LLVM 14)
ID: 7230 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7269 - Posted: 5 Nov 2022, 16:23:18 UTC

OpenCL 3.0.12 released (with his SDK)
ID: 7269 · Report as offensive    Reply Quote
Profile robertmiles

Send message
Joined: 13 Jan 09
Posts: 103
Credit: 331,865
RAC: 0
Message 7270 - Posted: 5 Nov 2022, 20:32:19 UTC

You could look for online introductory courses in OpenCL. I'm interested in such a course so I can start doing an OpenCL version for another BOINC project. Ralph@Home and Rosetta@Home look like very difficult first projects, though.

The online OpenCL courses I've found so far all appear to either be long and expensive, or be basically addons for someone who is already using OpenCL.
ID: 7270 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7288 - Posted: 17 Dec 2022, 10:11:22 UTC - in response to Message 7270.  
Last modified: 17 Dec 2022, 10:11:44 UTC

The online OpenCL courses I've found so far all appear to either be long and expensive, or be basically addons for someone who is already using OpenCL.


Today is not necessary to write code "directly" in OpenCl.
You can write c++ code and target OpenCl kernel and device.
For example, you can use oneAPI (it's SYCL) to target Nvidia and AMD gpu.
ID: 7288 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7292 - Posted: 29 Dec 2022, 15:32:06 UTC - in response to Message 7288.  

Today is not necessary to write code "directly" in OpenCl.
You can write c++ code and target OpenCl kernel and device.


An example: GPU performance portability using standard C++ with SYCL
ID: 7292 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7341 - Posted: 9 Apr 2023, 17:52:40 UTC

ID: 7341 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7364 - Posted: 22 Jun 2023, 10:10:18 UTC
Last modified: 22 Jun 2023, 10:10:48 UTC

PoCL 4.0 released, with support to OneApi driver (Intel GPU)
ID: 7364 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7365 - Posted: 22 Jun 2023, 10:22:38 UTC - in response to Message 7270.  

You could look for online introductory courses in OpenCL. I'm interested in such a course so I can start doing an OpenCL version for another BOINC project.


This is interesting, but it seems not for entry level user...
ID: 7365 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7390 - Posted: 21 Sep 2023, 9:07:05 UTC - in response to Message 7104.  

Kronos Group released updated documentation about C++ for OpenCL on github


The latest version is 3.0.14
ID: 7390 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7391 - Posted: 21 Sep 2023, 9:10:42 UTC
Last modified: 21 Sep 2023, 9:10:51 UTC

HypSycl/OpenSycl has renamed as AdaptiveCpp
ID: 7391 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7451 - Posted: 8 Dec 2023, 18:33:54 UTC

New PoCL 5.0 RC-1 has a new interesting features, OpenCl across networked system (over a LAN)
ID: 7451 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7470 - Posted: 3 Jan 2024, 19:01:57 UTC

ID: 7470 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7498 - Posted: 12 Apr 2024, 6:52:24 UTC - in response to Message 7470.  

ID: 7498 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7768 - Posted: 28 Jun 2024, 7:56:53 UTC - in response to Message 5025.  

Have you read the discussions over on Rosetta@home asking for Nvidia GPU support? The first efforts to try it showed that the minirosetta algorithms are too memory-demanding to make much use of enough GPU cores to get very much speedup on Nvidia boards with a currently reasonable amount of video memory. Therefore, such a conversion would require so much of a rewrite that it currently doesn't look reasonable.


This message is 14ys old, but it's true!!
Now we have a cuda/nvidia gpu app (or, better, we have a beta app)


Hope for an opencl/sycl/rocm/whatsoever app for other gpus (Amd, Intel, VideoCore, etc)!
ID: 7768 · Report as offensive    Reply Quote
Grant (SSSF)

Send message
Joined: 13 Jun 24
Posts: 124
Credit: 193,939
RAC: 2,635
Message 7809 - Posted: 16 Jul 2024, 10:33:32 UTC

New "SCALE" Software Allows Natively Compiling CUDA Apps For AMD GPUs
SCALE takes CUDA programs as-is and can even handle CUDA programs relying on line NVPTX Assembly. The SCALE compiler also is a drop-in replacement to NVIDIA's nvcc compiler and has a runtime that "impersonates" the NVIDIA CUDA Toolkit.

SCALE has been successfully tested with software like Blender, Llama-cpp, XGboost, FAISS, GOMC, STDGPU, Hashcat, and even NVIDIA Thrust. Spectral Compute has been testing SCALE across RDNA2 and RDNA3 GPUs along with basic testing on RDNA1 while Vega support is still a work-in-progress.

....

At its heart, SCALE is an nvcc-compatible compiler that can compile CUDA code for AMD GPUs, implementations of the CUDA runtime and driver APIs for AMD GPUs, and open-source wrapper libraries that in turn interface with AMD's ROCm libraries.

While ZLUDA for instance was quietly funded by AMD, I'm told by Spectral Compute that they've just been funding this development since 2017 via their consulting business. The only immediate downside seen with SCALE is that it itself is not open-source software but to at least there being a free edition license available for users.

Grant
Darwin NT
ID: 7809 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7810 - Posted: 17 Jul 2024, 7:39:12 UTC - in response to Message 7809.  
Last modified: 17 Jul 2024, 7:41:29 UTC

New "SCALE" Software Allows Natively Compiling CUDA Apps For AMD GPUs
[quote]SCALE takes CUDA programs as-is and can even handle CUDA programs relying on line NVPTX Assembly. The SCALE compiler also is a drop-in replacement to NVIDIA's nvcc compiler and has a runtime that "impersonates" the NVIDIA CUDA Toolkit.


There is A LOT of tools to convert automatically (or semi-automatically) CUDA code to OpenCL/OneApi/SYCL/etc.
From AdaptiveCpp (using C++ as common language between Cuda/OpenCl/Sycl) to SpectralCompute, from HIPIFY to Coriander
But there are 2 problems:
- Nvidia could block your work (as they did with ZLUDA). I know, it's not a good thing for science to block others competitor, but CUDA is closed, so Nvidia can do it.
- Developers have to "try and learn" new sw. And i don't know if they have the desire to....
ID: 7810 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7811 - Posted: 29 Jul 2024, 16:29:02 UTC - in response to Message 7810.  

There is A LOT of tools to convert automatically (or semi-automatically) CUDA code to OpenCL/OneApi/SYCL/etc.
From AdaptiveCpp (using C++ as common language between Cuda/OpenCl/Sycl) to SpectralCompute, from HIPIFY to Coriander


For example, AdaptiveCpp is faster than CUDA on Nvidia gpu!!!
ID: 7811 · Report as offensive    Reply Quote
Profile robertmiles

Send message
Joined: 13 Jan 09
Posts: 103
Credit: 331,865
RAC: 0
Message 7812 - Posted: 29 Jul 2024, 17:43:43 UTC - in response to Message 7811.  

There is A LOT of tools to convert automatically (or semi-automatically) CUDA code to OpenCL/OneApi/SYCL/etc.
From AdaptiveCpp (using C++ as common language between Cuda/OpenCl/Sycl) to SpectralCompute, from HIPIFY to Coriander


For example, AdaptiveCpp is faster than CUDA on Nvidia gpu!!
!


Where's your evidence that the Rosetta programs are written in a way that will allow such a conversion without using only graphics boards with extremely large numbers of GPU cores?
ID: 7812 · Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 9 Apr 08
Posts: 912
Credit: 1,892,541
RAC: 294
Message 7814 - Posted: 30 Jul 2024, 12:06:15 UTC - in response to Message 7812.  

Where's your evidence that the Rosetta programs are written in a way that will allow such a conversion without using only graphics boards with extremely large numbers of GPU cores?


No evidence.
Mine is only a little demonstration that is possible to avoid CUDA, if necessary...
ID: 7814 · Report as offensive    Reply Quote
Previous · 1 . . . 4 · 5 · 6 · 7 · 8 · Next

Message boards : Number crunching : Ralph support OpenCL ?



©2024 University of Washington
http://www.bakerlab.org