#oneapi — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #oneapi, aggregated by home.social.
-
Fanless AIM101 Edge System Integrates Intel Processor N150 and PCIe AI Accelerator Support
-
Even now, Thrust as a dependency is one of the main reason why we have a #CUDA backend, a #HIP / #ROCm backend and a pure #CPU backend in #GPUSPH, but not a #SYCL or #OneAPI backend (which would allow us to extend hardware support to #Intel GPUs). <https://doi.org/10.1002/cpe.8313>
This is also one of the reason why we implemented our own #BLAS routines when we introduced the semi-implicit integrator. A side-effect of this choice is that it allowed us to develop the improved #BiCGSTAB that I've had the opportunity to mention before <https://doi.org/10.1016/j.jcp.2022.111413>. Sometimes I do wonder if it would be appropriate to “excorporate” it into its own library for general use, since it's something that would benefit others. OTOH, this one was developed specifically for GPUSPH and it's tightly integrated with the rest of it (including its support for multi-GPU), and refactoring to turn it into a library like cuBLAS is
a. too much effort
b. probably not worth it.Again, following @eniko's original thread, it's really not that hard to roll your own, and probably less time consuming than trying to wrangle your way through an API that may or may not fit your needs.
6/
-
Even now, Thrust as a dependency is one of the main reason why we have a #CUDA backend, a #HIP / #ROCm backend and a pure #CPU backend in #GPUSPH, but not a #SYCL or #OneAPI backend (which would allow us to extend hardware support to #Intel GPUs). <https://doi.org/10.1002/cpe.8313>
This is also one of the reason why we implemented our own #BLAS routines when we introduced the semi-implicit integrator. A side-effect of this choice is that it allowed us to develop the improved #BiCGSTAB that I've had the opportunity to mention before <https://doi.org/10.1016/j.jcp.2022.111413>. Sometimes I do wonder if it would be appropriate to “excorporate” it into its own library for general use, since it's something that would benefit others. OTOH, this one was developed specifically for GPUSPH and it's tightly integrated with the rest of it (including its support for multi-GPU), and refactoring to turn it into a library like cuBLAS is
a. too much effort
b. probably not worth it.Again, following @eniko's original thread, it's really not that hard to roll your own, and probably less time consuming than trying to wrangle your way through an API that may or may not fit your needs.
6/
-
Even now, Thrust as a dependency is one of the main reason why we have a #CUDA backend, a #HIP / #ROCm backend and a pure #CPU backend in #GPUSPH, but not a #SYCL or #OneAPI backend (which would allow us to extend hardware support to #Intel GPUs). <https://doi.org/10.1002/cpe.8313>
This is also one of the reason why we implemented our own #BLAS routines when we introduced the semi-implicit integrator. A side-effect of this choice is that it allowed us to develop the improved #BiCGSTAB that I've had the opportunity to mention before <https://doi.org/10.1016/j.jcp.2022.111413>. Sometimes I do wonder if it would be appropriate to “excorporate” it into its own library for general use, since it's something that would benefit others. OTOH, this one was developed specifically for GPUSPH and it's tightly integrated with the rest of it (including its support for multi-GPU), and refactoring to turn it into a library like cuBLAS is
a. too much effort
b. probably not worth it.Again, following @eniko's original thread, it's really not that hard to roll your own, and probably less time consuming than trying to wrangle your way through an API that may or may not fit your needs.
6/
-
Even now, Thrust as a dependency is one of the main reason why we have a #CUDA backend, a #HIP / #ROCm backend and a pure #CPU backend in #GPUSPH, but not a #SYCL or #OneAPI backend (which would allow us to extend hardware support to #Intel GPUs). <https://doi.org/10.1002/cpe.8313>
This is also one of the reason why we implemented our own #BLAS routines when we introduced the semi-implicit integrator. A side-effect of this choice is that it allowed us to develop the improved #BiCGSTAB that I've had the opportunity to mention before <https://doi.org/10.1016/j.jcp.2022.111413>. Sometimes I do wonder if it would be appropriate to “excorporate” it into its own library for general use, since it's something that would benefit others. OTOH, this one was developed specifically for GPUSPH and it's tightly integrated with the rest of it (including its support for multi-GPU), and refactoring to turn it into a library like cuBLAS is
a. too much effort
b. probably not worth it.Again, following @eniko's original thread, it's really not that hard to roll your own, and probably less time consuming than trying to wrangle your way through an API that may or may not fit your needs.
6/
-
Even now, Thrust as a dependency is one of the main reason why we have a #CUDA backend, a #HIP / #ROCm backend and a pure #CPU backend in #GPUSPH, but not a #SYCL or #OneAPI backend (which would allow us to extend hardware support to #Intel GPUs). <https://doi.org/10.1002/cpe.8313>
This is also one of the reason why we implemented our own #BLAS routines when we introduced the semi-implicit integrator. A side-effect of this choice is that it allowed us to develop the improved #BiCGSTAB that I've had the opportunity to mention before <https://doi.org/10.1016/j.jcp.2022.111413>. Sometimes I do wonder if it would be appropriate to “excorporate” it into its own library for general use, since it's something that would benefit others. OTOH, this one was developed specifically for GPUSPH and it's tightly integrated with the rest of it (including its support for multi-GPU), and refactoring to turn it into a library like cuBLAS is
a. too much effort
b. probably not worth it.Again, following @eniko's original thread, it's really not that hard to roll your own, and probably less time consuming than trying to wrangle your way through an API that may or may not fit your needs.
6/
-
I'm getting the material ready for my upcoming #GPGPU course that starts on March. Even though I most probably won't get to it,I also checked my trivial #SYCL programs. Apparently the 2025.0 version of the #Intel #OneAPI #DPCPP runtime doesn't like any #OpenCL platform except Intel's own (I have two other platforms that support #SPIRV, so why aren't they showing up? From the documentation I can find online this should be sufficient, but apparently it's not …)
-
I'm getting the material ready for my upcoming #GPGPU course that starts on March. Even though I most probably won't get to it,I also checked my trivial #SYCL programs. Apparently the 2025.0 version of the #Intel #OneAPI #DPCPP runtime doesn't like any #OpenCL platform except Intel's own (I have two other platforms that support #SPIRV, so why aren't they showing up? From the documentation I can find online this should be sufficient, but apparently it's not …)
-
I'm getting the material ready for my upcoming #GPGPU course that starts on March. Even though I most probably won't get to it,I also checked my trivial #SYCL programs. Apparently the 2025.0 version of the #Intel #OneAPI #DPCPP runtime doesn't like any #OpenCL platform except Intel's own (I have two other platforms that support #SPIRV, so why aren't they showing up? From the documentation I can find online this should be sufficient, but apparently it's not …)
-
I'm getting the material ready for my upcoming #GPGPU course that starts on March. Even though I most probably won't get to it,I also checked my trivial #SYCL programs. Apparently the 2025.0 version of the #Intel #OneAPI #DPCPP runtime doesn't like any #OpenCL platform except Intel's own (I have two other platforms that support #SPIRV, so why aren't they showing up? From the documentation I can find online this should be sufficient, but apparently it's not …)
-
I'm getting the material ready for my upcoming #GPGPU course that starts on March. Even though I most probably won't get to it,I also checked my trivial #SYCL programs. Apparently the 2025.0 version of the #Intel #OneAPI #DPCPP runtime doesn't like any #OpenCL platform except Intel's own (I have two other platforms that support #SPIRV, so why aren't they showing up? From the documentation I can find online this should be sufficient, but apparently it's not …)
-
Just how deep is #Nvidia's #CUDA moat really?
Not as impenetrable as you might think, but still more than Intel or AMD would like
It's not enough just to build a competitive part: you also have to have #software that can harness all those #FLOPS — something Nvidia has spent the better part of two decades building with its CUDA runtime, while competing frameworks for low-level #GPU #programming are far less mature like AMD's #ROCm or Intel's #OneAPI.
https://www.theregister.com/2024/12/17/nvidia_cuda_moat/ #developers -
Just how deep is #Nvidia's #CUDA moat really?
Not as impenetrable as you might think, but still more than Intel or AMD would like
It's not enough just to build a competitive part: you also have to have #software that can harness all those #FLOPS — something Nvidia has spent the better part of two decades building with its CUDA runtime, while competing frameworks for low-level #GPU #programming are far less mature like AMD's #ROCm or Intel's #OneAPI.
https://www.theregister.com/2024/12/17/nvidia_cuda_moat/ #developers -
Just how deep is #Nvidia's #CUDA moat really?
Not as impenetrable as you might think, but still more than Intel or AMD would like
It's not enough just to build a competitive part: you also have to have #software that can harness all those #FLOPS — something Nvidia has spent the better part of two decades building with its CUDA runtime, while competing frameworks for low-level #GPU #programming are far less mature like AMD's #ROCm or Intel's #OneAPI.
https://www.theregister.com/2024/12/17/nvidia_cuda_moat/ #developers -
Just how deep is #Nvidia's #CUDA moat really?
Not as impenetrable as you might think, but still more than Intel or AMD would like
It's not enough just to build a competitive part: you also have to have #software that can harness all those #FLOPS — something Nvidia has spent the better part of two decades building with its CUDA runtime, while competing frameworks for low-level #GPU #programming are far less mature like AMD's #ROCm or Intel's #OneAPI.
https://www.theregister.com/2024/12/17/nvidia_cuda_moat/ #developers -
Just how deep is #Nvidia's #CUDA moat really?
Not as impenetrable as you might think, but still more than Intel or AMD would like
It's not enough just to build a competitive part: you also have to have #software that can harness all those #FLOPS — something Nvidia has spent the better part of two decades building with its CUDA runtime, while competing frameworks for low-level #GPU #programming are far less mature like AMD's #ROCm or Intel's #OneAPI.
https://www.theregister.com/2024/12/17/nvidia_cuda_moat/ #developers -
I should mention that this isn't just a matter of the dominant player intentionally boycotting standards that would make them lose the vendor lock-in advantage (hello #NVIDIA). All major vendors are guilty of this one way or the other. For example, #AMD unjustifiably pulled (or maybe failed to add) #SPIR and #CPU support from their new #OpenCL implementation. #Intel's #oneAPI (even while still leveraging the OpenCL backend) effectively failed on any other OpenCL platform.
-
I should mention that this isn't just a matter of the dominant player intentionally boycotting standards that would make them lose the vendor lock-in advantage (hello #NVIDIA). All major vendors are guilty of this one way or the other. For example, #AMD unjustifiably pulled (or maybe failed to add) #SPIR and #CPU support from their new #OpenCL implementation. #Intel's #oneAPI (even while still leveraging the OpenCL backend) effectively failed on any other OpenCL platform.
-
I should mention that this isn't just a matter of the dominant player intentionally boycotting standards that would make them lose the vendor lock-in advantage (hello #NVIDIA). All major vendors are guilty of this one way or the other. For example, #AMD unjustifiably pulled (or maybe failed to add) #SPIR and #CPU support from their new #OpenCL implementation. #Intel's #oneAPI (even while still leveraging the OpenCL backend) effectively failed on any other OpenCL platform.
-
.@Intel Advanced Matrix Extensions [AMX] Performance With Xeon Scalable #SapphireRapids
-- The big #AI performance uplift and power efficiency benefits from #AMX w/ #oneAPI #oneDNN & #OpenVINO benchmarks
https://www.phoronix.com/review/intel-xeon-amx
Original tweet : https://twitter.com/phoronix/status/1615054211582164993
-
Howdy all - registrations are still open for the first oneAPI DevSummit hosted by the UXL Foundation! Learn about GPGPU programming, oneAPI and how companies are coalescing around #oneapi / #sycl
https://linuxfoundation.regfox.com/oneapiuxldevsummit2024Registration will closeat 5pm today. The DevSummit will start at 8pm PT or 8:30am IST. See you there!
-
📢 Introduction to #oneAPI, #SYCL2020 & #OpenMP offloading
📆September 23-25, 2024In this 3-day online course, HLRS - High-Performance Computing Center Stuttgart provides an introduction to Intel Corporation's oneAPI implementation 🖥
Read more & Register👉 https://www.hlrs.de/training/2024/intel-oneapi
-
Just one more day to submit your session for the UXL oneAPI DevSummit being held October 9th & 10th!
Learn more: https://sessionize.com/uxldevsummit
#SYCL #oneAPI #UXL -
#oneAPI / #SYCL 2020 / #DPC++ / #AdaptiveCpp Is starting to look pretty nice. I’m not sure if there is a story for mobile devices and older devices. It seems that OpenCL 1.2 isn’t enough for SYCL 2020, but unsure. Also not sure how Vulcan fits in here. It’s also unclear how it scales to smaller kernels and data. The API looks like kernel/buffer/queue. Not sure if you can queue up many kernels or if they can be pipelined.
How does the new IR that llvm is using fit in here? So many questions.
-
-
A coalition led by Qualcomm, Google, and Intel, under the UXL Foundation, aims to break $2.2 trillion Nvidia's stronghold on the AI market by developing an open-source software suite that supports diverse AI accelerator chips, leveraging Intel's OneAPI. #nvidia #google #intel #Qualcomm #ai #opensource #chips #semiconductor #strategy #partnership #oneapi #api #engineer #engineering #software #market
-
#PoCL 4.0 @openclapi Implementation Released With @IntelGraphics #oneAPI Level Zero Driver
https://www.phoronix.com/news/PoCL-4.0-Released
Original tweet : https://twitter.com/phoronix/status/1671820647540957185
-
RT from Software Dev Tools (@IntelDevTools)
Take advantage of performance-optimized Intel 4th Gen #IntelXeon processors and Advanced Matrix Extensions [AMX] Performance for all your #AI and machine learning related workloads. #ML #oneAPIhttps://twitter.com/phoronix/status/1615069457260445697
Original tweet : https://twitter.com/IntelDevTools/status/1615132356410740736
-
Intel’s run at the GPU market begins with Tiger Lake onboard graphics - Enlarge / Intel is looking to replace Nvidia as the "one stop GPU shop," with a comprehensive line ... - https://arstechnica.com/?p=1699348 #uncategorized #deeplearning #laptopgpu #intelxe #oneapi #intel #tech #gpu
-
#oneAPI / #SYCL 2020 / #DPC++ / #AdaptiveCpp Is starting to look pretty nice. I’m not sure if there is a story for mobile devices and older devices. It seems that OpenCL 1.2 isn’t enough for SYCL 2020, but unsure. Also not sure how Vulcan fits in here. It’s also unclear how it scales to smaller kernels and data. The API looks like kernel/buffer/queue. Not sure if you can queue up many kernels or if they can be pipelined.
How does the new IR that llvm is using fit in here? So many questions.
-
#oneAPI / #SYCL 2020 / #DPC++ / #AdaptiveCpp Is starting to look pretty nice. I’m not sure if there is a story for mobile devices and older devices. It seems that OpenCL 1.2 isn’t enough for SYCL 2020, but unsure. Also not sure how Vulcan fits in here. It’s also unclear how it scales to smaller kernels and data. The API looks like kernel/buffer/queue. Not sure if you can queue up many kernels or if they can be pipelined.
How does the new IR that llvm is using fit in here? So many questions.
-
#oneAPI / #SYCL 2020 / #DPC++ / #AdaptiveCpp Is starting to look pretty nice. I’m not sure if there is a story for mobile devices and older devices. It seems that OpenCL 1.2 isn’t enough for SYCL 2020, but unsure. Also not sure how Vulcan fits in here. It’s also unclear how it scales to smaller kernels and data. The API looks like kernel/buffer/queue. Not sure if you can queue up many kernels or if they can be pipelined.
How does the new IR that llvm is using fit in here? So many questions.
-
#oneAPI / #SYCL 2020 / #DPC++ / #AdaptiveCpp Is starting to look pretty nice. I’m not sure if there is a story for mobile devices and older devices. It seems that OpenCL 1.2 isn’t enough for SYCL 2020, but unsure. Also not sure how Vulcan fits in here. It’s also unclear how it scales to smaller kernels and data. The API looks like kernel/buffer/queue. Not sure if you can queue up many kernels or if they can be pipelined.
How does the new IR that llvm is using fit in here? So many questions.
-
Das Programmiermodell zum Entwickeln von Anwendungen auf heterogenen Systemen nimmt mit der 1.0-Relase der Spezifikation endgültig Gestalt an.
Heterogene Systeme: Intel veröffentlicht oneAPI-Spezifikation 1.0 -
We have a meetup today - if anyone want to spend a pleasant hour talking about oneAPI AI toolkits - my colleague Edwardo Alvarez will doing a free wheeling chat about these kits and honestly, we'll just be winging it about other stuff as well. You can join us at https://meetup.com/oneapi-community-us (to join) or if you want to bypass that shoot me an DM and I'll give you the video link:
TODAY at 11:30am PT - thank you! #hpc #AI #artificialintelligence #oneapi #iamintel #highperformancecomputing #randomhashtag -
I just finished my intro and we have Andres Rodriguez speaking right now about AI through oneAPI #intel #oneapi #DevSummit #iamintel
-
Next week is our oneAPI DevSummit for AI and HPC - have you registered yet??! We have at least one hands-on demo that's worth checking out!
https://www.oneapi.io/events/oneapi-devsummit-for-ai-and-hpc-2022/
#hpc #ai #oneapi #iamintel -
Hey folks - the conference I am help organizing (no, not LinuxAppSummit) for #oneAPI is now open for registration - please check out virtual oneAPI DevSummit for #AI and #HPC check us out at https://www.oneapi.io/events/oneapi-devsummit-for-ai-and-hpc-2022/
#oneapi #iamintel #sycl #artificialintelligence #highperformancecomputing -
Hello #HPC, #AI, #Machinelearning folks! Just wanted to let everyone know that registrations are open for the oneAPI DevSummit - looking forward to seeing some of you virtually there! It's an online summit which means that everyone will be able to attend regardless of where in the world you are. :)
https://www.oneapi.io/events/oneapi-devsummit-for-ai-and-hpc-2022/
Shoot me any questions you have - the rest of y'alls can ignore these posts if you filter out #oneapi 🙂
-
work related - if you don't want to see that part of the feed - make sure to filter out #oneapi :-)
So, I worked on this blog post on #oneAPI - and hopefully my #hpc and #ai peeps will weigh in on it! (either here or on dev.to) - being able to multi-architecture programming is a good thing rather than restricting yourself to only one platform. Love to know the feedback.
My next blog post is going to be how do we update the #sycl spec?!
-
.@IntelDevTools Releases #oneAPI #oneDNN 3.0 In Advance Of #SapphireRapids
-- Plus optimizations for other Intel CPUs & GPUs, including initial Granite Rapids support.
https://www.phoronix.com/news/Intel-oneDNN-3.0
Original tweet : https://twitter.com/phoronix/status/1605147559525335041
-
#Intel #oneAPI @IntelDevTools @IntelSoftware #oneDNN 3.0 Being Prepared With More Performance Optimizations
-- Plus optimizations for Arm, Power, NVIDIA & AMD hardware too.
https://www.phoronix.com/news/Intel-oneDNN-3.0-RC
Original tweet : https://twitter.com/phoronix/status/1599012110301679616
-
I'll be speaking during the OneAPI dev summit tomorrow, specifically the panel discussion on accelerated computing.
Partially as a meme/Sanity check the challenge of this afternoon:
Can I got from no installs to a GPU accelerated simulation of OpenFOAM using an Arc a770 on Linux?
Plan is #OpenFoam 2212, #petsc 3.19, #OneAPI 2023.1 and #mesa 23 (in case I need to fall back to #OpenCL)
-
#BSI WID-SEC-2024-3422: [NEU] [mittel] #Intel #oneAPI #Math #Kernel #Library: Schwachstelle ermöglicht Privilegieneskalation
Ein lokaler Angreifer kann eine Schwachstelle in Intel oneAPI Math Kernel Library ausnutzen, um seine Privilegien zu erhöhen.
https://wid.cert-bund.de/portal/wid/securityadvisory?name=WID-SEC-2024-3422
-
#BSI WID-SEC-2024-3422: [NEU] [mittel] #Intel #oneAPI #Math #Kernel #Library: Schwachstelle ermöglicht Privilegieneskalation
Ein lokaler Angreifer kann eine Schwachstelle in Intel oneAPI Math Kernel Library ausnutzen, um seine Privilegien zu erhöhen.
https://wid.cert-bund.de/portal/wid/securityadvisory?name=WID-SEC-2024-3422