HPC Guru (Twitter)
Analyzing performance profiles using #Hatchet, an #OpenSource #Python-based library started at @Livermore_Lab
Hatchet facilitates comparing performance profiles from multiple executions
@fosdem_hpc talk by @bhatele at @hpc_group
https://www.youtube.com/watch?v=VcJ3BPJCc18
#HPC #fosdem21
Analyzing performance profiles using #Hatchet, an #OpenSource #Python-based library started at @Livermore_Lab
Hatchet facilitates comparing performance profiles from multiple executions
@fosdem_hpc talk by @bhatele at @hpc_group
https://www.youtube.com/watch?v=VcJ3BPJCc18
#HPC #fosdem21
YouTube
Analyzing Performance Profiles Using Hatchet
Presented at FOSDEM '21 HPC Devroom
Performance analysis is critical for identifying and eliminating bottlenecks in both serial and parallel programs. There are many profiling tools that can instrument serial and parallel codes, and gather performance data.…
Performance analysis is critical for identifying and eliminating bottlenecks in both serial and parallel programs. There are many profiling tools that can instrument serial and parallel codes, and gather performance data.…
HPC Guru (Twitter)
#OGHPC Panel: @DanStanzione sees a split between the #Exascale crowd and everyone else in #HPC
The "everyone else" group wants #Python!
#OGHPC Panel: @DanStanzione sees a split between the #Exascale crowd and everyone else in #HPC
The "everyone else" group wants #Python!
HPC Guru (Twitter)
#Python turns 30: A Q&A with Pablo Galindo (@pyblogsal), a member of the Python Steering Council
https://www.techradar.com/uk/news/python-turns-30-meet-the-man-tasked-with-keeping-the-programming-language-on-track
#Python turns 30: A Q&A with Pablo Galindo (@pyblogsal), a member of the Python Steering Council
https://www.techradar.com/uk/news/python-turns-30-meet-the-man-tasked-with-keeping-the-programming-language-on-track
NVIDIA (Twitter)
RT @NVIDIAHPCDev: Announcing CUDA Python, another step towards simplifying the developer experience with improved #Python code portability and compatibility: https://nvda.ws/3dOn5S7 #GTC21 https://twitter.com/NVIDIAHPCDev/status/1382438514848063490/photo/1
RT @NVIDIAHPCDev: Announcing CUDA Python, another step towards simplifying the developer experience with improved #Python code portability and compatibility: https://nvda.ws/3dOn5S7 #GTC21 https://twitter.com/NVIDIAHPCDev/status/1382438514848063490/photo/1
HPC Guru (Twitter)
Testing #Julia: Fast as #Fortran, Beautiful as #Python - by @MartinDMaas
https://www.matecdev.com/posts/numpy-julia-fortran.html
@JuliaLanguage #HPC via @dataelixir @hpcjoe
Testing #Julia: Fast as #Fortran, Beautiful as #Python - by @MartinDMaas
https://www.matecdev.com/posts/numpy-julia-fortran.html
@JuliaLanguage #HPC via @dataelixir @hpcjoe
HPC Guru (Twitter)
RT @LorenaABarba: We edited this special issue with Andreas Kloeckner, @prabhu_r and @r_c_thomas, and it is jam-packed of good stuff—check it out!
#Python #HPC
RT @LorenaABarba: We edited this special issue with Andreas Kloeckner, @prabhu_r and @r_c_thomas, and it is jam-packed of good stuff—check it out!
#Python #HPC
HPC Guru (Twitter)
RT @thoefler: If you missed the COLOC and Russian Supercomputing Days keynotes, then you can find a recording at https://www.youtube.com/watch?v=iZlhU6fS3cU now.
Data-Centric #Python for #HPC in the wild in 25 minutes! Full code and information at https://github.com/spcl/dace
RT @thoefler: If you missed the COLOC and Russian Supercomputing Days keynotes, then you can find a recording at https://www.youtube.com/watch?v=iZlhU6fS3cU now.
Data-Centric #Python for #HPC in the wild in 25 minutes! Full code and information at https://github.com/spcl/dace
YouTube
Data-Centric Python - Productivity, portability and all with high performance!
Keynote talk at COLOC/EuroPar'21
HPC Guru (Twitter)
#Python has become a major language for #HPC/#AI development, ahead of #Fortran
@addisonsnell at Swiss @hpccouncil
#Python has become a major language for #HPC/#AI development, ahead of #Fortran
@addisonsnell at Swiss @hpccouncil
HPCwire (Twitter)
Solving Heterogeneous Programming Challenges with Python, Today - by @JamesReinders
http://ow.ly/uFxy50JLFyC #HPC #Python
Solving Heterogeneous Programming Challenges with Python, Today - by @JamesReinders
http://ow.ly/uFxy50JLFyC #HPC #Python
HPCwire
Solving Heterogeneous Programming Challenges with Python, Today
You may be surprised how ready Python is for heterogeneous programming, and how easy it is to use today. Our first three articles about heterogeneous programming focused primarily on C++ as we ponder “how to enable programming in the face of an explosion…
HPC Guru (Twitter)
👇 @tgamblin on the lack of representation for #HPC or scientific computing in the 2022 #Python Dev Survey
-----------
@tgamblin:
.@jetbrains @thepsf: Kinda sad not to see scientific computing/HPC represented in the 2022 Python Dev Survey.
- No Fortran option for "other languages"
- No Scientific Computing or HPC options for field
Scientific Computing isn't just data analysis/ML.
https://t.co/h1cQgCFvBM
👇 @tgamblin on the lack of representation for #HPC or scientific computing in the 2022 #Python Dev Survey
-----------
@tgamblin:
.@jetbrains @thepsf: Kinda sad not to see scientific computing/HPC represented in the 2022 Python Dev Survey.
- No Fortran option for "other languages"
- No Scientific Computing or HPC options for field
Scientific Computing isn't just data analysis/ML.
https://t.co/h1cQgCFvBM
HPCwire (Twitter)
Do not miss our “interviews” with #ChatGPT and #Eliza on topics that span trusting #AI, #HPC, heterogeneous #computing, #Fortran, #Python, and more. http://ow.ly/oeWq50MSc3v
Do not miss our “interviews” with #ChatGPT and #Eliza on topics that span trusting #AI, #HPC, heterogeneous #computing, #Fortran, #Python, and more. http://ow.ly/oeWq50MSc3v
HPC Guru (Twitter)
RT @GeorgHager: "The Missing Python Introduction for Scientists" - great talk on efficiency in #Python by physicist Tamás Gál at the NHR@FAU #HPC Café https://youtu.be/maPozWKuv98 https://twitter.com/GeorgHager/status/1640351796752261123/photo/1
RT @GeorgHager: "The Missing Python Introduction for Scientists" - great talk on efficiency in #Python by physicist Tamás Gál at the NHR@FAU #HPC Café https://youtu.be/maPozWKuv98 https://twitter.com/GeorgHager/status/1640351796752261123/photo/1
HPC Guru (Twitter)
MIT Turbocharges #Python’s Notoriously Slow Compiler
#Codon lets users run #Python code as efficiently as C or C++
https://spectrum.ieee.org/python-compiler
#HPC #AI via @IEEESpectrum
MIT Turbocharges #Python’s Notoriously Slow Compiler
#Codon lets users run #Python code as efficiently as C or C++
https://spectrum.ieee.org/python-compiler
#HPC #AI via @IEEESpectrum
HPC Guru (Twitter)
CFP: High Performance Python for Science at Scale (HPPSS)
HPPSS aims to bring together the #Python user community attempting to scale code for #HPC, #AI and Data for science
The workshop will take place at #SC23 Sun morning Nov 12
https://hppss.github.io/SC23/callfordemos.html
via @CosmoCa3sar
CFP: High Performance Python for Science at Scale (HPPSS)
HPPSS aims to bring together the #Python user community attempting to scale code for #HPC, #AI and Data for science
The workshop will take place at #SC23 Sun morning Nov 12
https://hppss.github.io/SC23/callfordemos.html
via @CosmoCa3sar
HPC Guru (Twitter)
Nvidia’s Jensen Huang: #AI is ushering in the era of Software 3.0
Creating & running applications will be as simple as writing queries into a universal AI translator, running a few lines of #Python code, and selecting an AI model of your choice
https://www.enterpriseai.news/2023/06/28/nvidia-ceo-huang-get-ready-for-software-3-0/
Nvidia’s Jensen Huang: #AI is ushering in the era of Software 3.0
Creating & running applications will be as simple as writing queries into a universal AI translator, running a few lines of #Python code, and selecting an AI model of your choice
https://www.enterpriseai.news/2023/06/28/nvidia-ceo-huang-get-ready-for-software-3-0/
HPC Guru (Twitter)
Is #Mojo the #Fortran for #AI programming, or more?
One of the big changes between #Python and Mojo is that Mojo is multithreaded and can run across multiple cores - Modular AI show a 35,000X speedup calculating Mandelbrot sets
nextplatform.com/2023/09/08/…
#HPC via @TDaytonPM
Is #Mojo the #Fortran for #AI programming, or more?
One of the big changes between #Python and Mojo is that Mojo is multithreaded and can run across multiple cores - Modular AI show a 35,000X speedup calculating Mandelbrot sets
nextplatform.com/2023/09/08/…
#HPC via @TDaytonPM