HPC & Quantum
26 subscribers
11.4K photos
668 videos
3 files
30.6K links
Download Telegram
HPC Guru (Twitter)

RT @IntelSoftware: Universidad de Málaga migrated the Tsunami-HySEA (Hyperbolic Systems and Efficient Algorithms) prediction simulator from #CUDA to #SYCL, allowing users to be independent of hardware vendors. See this and more in our CUDA to SYCL catalog. intel.ly/3TyvK1t
HPC Guru (Twitter)

RT @NVIDIAHPCDev: Our NVIDIA Compute Sanitizer is a powerful tool that can save you time and effort while improving the reliability and performance of your #CUDA applications.

🐛 Learn how to efficiently debug code in the CUDA environment:

nvda.ws/3VuY1Hs
HPC Guru (Twitter)

Even as they continue to spend billions of dollars on @nvidia’s latest products, big tech companies hope that #Triton will help break the stranglehold that the chipmaker has on #AI hardware

https://www.ft.com/content/320f35de-9a6c-4dbf-b42f-9cdaf35e45bb

#HPC #GPU #CUDA via @FT
HPC Guru (Twitter)

RT @NVIDIAHPCDev: Explore the new updates in #HPC SDK v24.5, optimized with support for the #NVPL components and #CUDA 12.4.

➡️ nvda.ws/4e2tjwe
HPC Guru (Twitter)

💯 Some people underestimate the contribution of @nvidia software in their success

It’s not just having great #GPU & networking silicon, it’s the libraries and software tools/packages that @NVIDIAHPCDev & @NVIDIAAI invested in that makes them so valuable

#AI #HPC #CUDA https://twitter.com/matteastwood/status/1803170389096321399#m
HPCwire (Twitter)

As AI continues to expand, the choice of software stacks like #CUDA, #ROCm, and #oneAPI becomes crucial for performance and productivity in #GPU-centric tasks. Our latest article delves into these software stacks and explores higher-level #HPC languages. ow.ly/7v5h50SpwYE
HPC Guru (Twitter)

Checkpoint and restore functionality for #CUDA is exposed via a command-line utility called cuda-checkpoint

Combine it with #CRIU (Checkpoint/Restore in Userspace), an open-source checkpointing utility, to fully checkpoint CUDA apps

https://developer.nvidia.com/blog/checkpointing-cuda-applications-with-criu/

#HPC #AI #GPU
HPC Guru (Twitter)

.@RajaXg:

For architects aspiring to build something new:

The demand for compute and bandwidth is increasing 3-4X every year

Based on first principles laid out here, #CUDA #GPU hardware will be disrupted

Only question is by "who" and "when"?

#AI #HPC
HPC Guru (Twitter)

RT @NVIDIADC: 📣 CUDA MODE goes live IRL! Join us in San Francisco on Sept. 21 for the #Hackathon event.

🎉 Make friends, enjoy keynotes from #CUDA experts, then hack all day and night!

➡️ Apply for a spot now: nvda.ws/3AcLAHB
HPC Guru (Twitter)

RT @NVIDIAHPCDev: 📢 Announcing the #CUDA Toolkit, release 12.6

Supports GB100 capabilities
New library enhancements to cuBLAS, cuFFT, cuSOLVER, cuSPARSE
Release of Nsight Compute 2024.3

📥 nvda.ws/3yiraMW
HPC Guru (Twitter)

ParaS is an implementation of the SYCL 2020 specification, developed for architectural and vendor neutral programming

The current version enables seamless execution across CPUs of #x86 (Intel and AMD) and #ARM as well as #CUDA enabled NVIDIA GPUs

#HPC #AI via @cdacindia https://twitter.com/cdacindia/status/1840063621981700424#m
HPC Guru (Twitter)

Peachy Assignments serve as effective tools for #teaching specific programming models and provide insights into the portability of concepts and techniques across different models

https://gamuva.infor.uva.es/peachy-assignments/

#HPC #SC24 #MPI #CUDA #OpenCL #OpenMP
via @jgarciames.bsky.social
HPC Guru (Twitter)

RT @NVIDIAHPCDev: 🌌 @Google Quantum AI is performing the largest #quantum simulations using our NVIDIA #CUDA-Q functionality. Learn how dynamics simulations take advantage of the scalability offered by multi-GPU capabilities. #SC24

➡️ nvda.ws/3AMR2Si
HPC Guru (Twitter)

RT @NVIDIAHPCDev: 📣 Just released: NVIDIA Warp 1.5, a new #opensource #Python framework for #CUDA acceleration.

➡️ nvda.ws/3BpJ1Tq

Accelerate the performance of your scientific computing workflows, enabling quicker simulations in robotics, fluid dynamics, and digital twins.
HPC Guru (Twitter)

.@deepseek_ai's breakthrough was achieved by implementing tons of fine-grained optimizations and usage of assembly-like PTX programming instead of @nvidia's #CUDA, according to an analysis from Mirae Asset Securities Korea cited by @Jukanlosreve

https://www.tomshardware.com/tech-industry/artificial-intelligence/deepseeks-ai-breakthrough-bypasses-industry-standard-cuda-uses-assembly-like-ptx-programming-instead

#AI
HPC Guru (Twitter)

RT @hpcnotes: What language will be most used to write new #HPC applications/code over the next few years?

#Fortran?

C/C++?

Julia?

Python?

Copilot / ChatGPT / other #AI prompt languages?

#CUDA?