Chapel Parallel Programming Language
Chapel Parallel Programming Language
  • Видео 112
  • Просмотров 12 387
The Game of Life: A multi-GPU implementation in Chapel
Conway's Game of Life is a classic example of a cellular automaton. In this video, we demonstrate how to implement the Game of Life in Chapel and run it on a GPU. The result is a portable and readable implementation that can run on any system with or without a GPU.
Check out the code at: github.com/jabraham17/hpe-dev-meetup-chapel-july-2024/blob/main/05_life.chpl
This demo was part of larger talk about GPU programming in Chapel, available here: ruclips.net/video/nj-WqhGEy24/видео.html
A playlist containing all five demos is available here: ruclips.net/p/PLuqM5RJ2KYFgNM62w9QdRTcWPpf3rfSwr
Просмотров: 35

Видео

Foundations of AI: A multi-GPU softmax function in Chapel
Просмотров 2016 часов назад
AI relies on optimized math done on GPUs. In this video, we demonstrate how the crucial softmax function can be implemented in Chapel and run on many GPUs. We also show off more of Chapel's parallel-first language features, which are essential to writing portable, readable, and maintainable GPU code. Check out the code at: github.com/jabraham17/hpe-dev-meetup-chapel-july-2024/blob/main/04_softm...
Foundations of AI: A multi-GPU sigmoid function in Chapel
Просмотров 1816 часов назад
AI relies on optimized math done on GPUs. In this video, we demonstrate how the crucial sigmoid function can be implemented in Chapel and run on many GPUs. Check out the code at: github.com/jabraham17/hpe-dev-meetup-chapel-july-2024/blob/main/03_sigmoid.chpl This demo was part of larger talk about GPU programming in Chapel, available here: ruclips.net/video/nj-WqhGEy24/видео.html A playlist con...
Vendor-neutral multi-GPU programming in Chapel
Просмотров 3616 часов назад
See how to target multiple GPUs from the same program in Chapel. In this video, we demonstrate how Chapel programs can target as many GPUs as are available on the system, regardless of if they are from AMD or NVIDIA. Check out the code at: github.com/jabraham17/hpe-dev-meetup-chapel-july-2024/blob/main/02_arrayAssign.chpl This demo was part of larger talk about GPU programming in Chapel, availa...
Simple GPU programming in Chapel
Просмотров 5416 часов назад
See how to write a simple GPU program in Chapel. In this video, we demonstrate how Chapel's parallel-first language features make portable GPU programming easier. Check out the code at: github.com/jabraham17/hpe-dev-meetup-chapel-july-2024/blob/main/01_simpleSingleGpu.chpl This demo was part of larger talk about GPU programming in Chapel, available here: ruclips.net/video/nj-WqhGEy24/видео.html...
Getting Started with Chapel: Functions, Loops, and Arrays
Просмотров 76Месяц назад
This is a live recording of Michael Ferguson's Chapel demo session, recorded live on July 11, 2024. To see other upcoming demos, refer to the schedule at chapel-lang.org/events.html
The Value of Languages in Parallel Computing
Просмотров 157Месяц назад
This is Brad Chamberlain's lightning talk from PNW PLSE 2024 (pnwplse.org/), recorded live on May 7, 2024 at the University of Washington in Seattle, WA. Its full title is "The Value of Languages in Parallel Computing: Aggregated vs. Non-Aggregated Communication in Distributed Computing Settings." The unabridged version of the talk's slide deck is available here: chapel-lang.org/presentations/C...
Arkouda Tutorial for ChapelCon '24
Просмотров 110Месяц назад
This is a recording of the Arkouda Tutorial presented at ChapelCon '24. Slides are available at chapel-lang.org/ChapelCon/2024/arkouda-tutorial.pdf Codespace used in this tutorial: github.com/bmcdonald3/chapelcon-2024-arkouda Arkouda repository: github.com/bears-r-us/arkouda Arkouda documentation: bears-r-us.github.io/arkouda/
Chapel Tutorial from ChapelCon '24
Просмотров 198Месяц назад
This is a recording of the Chapel Tutorial presented at ChapelCon '24. Slides are available at chapel-lang.org/ChapelCon/2024/chapel-tutorial.pdf * 00:00 - What is Chapel? * 03:40 - The Heat Diffusion Problem * 08:00 - Chapel code to Simulate Heat Diffusion * 12:45 - Hands-on #1 (getting the code) * 15:58 - Hands-on #2 (varying problem size) * 16:52 - Hands-on #3 (the fast flag) * 19:26 - Langu...
ChapelCon '24: Follow-Up on Chapel-Powered HPC Workflows for Python
Просмотров 24Месяц назад
This is Harumi Kuno's ChapelCon'24 talk, recorded live on June 7, 2024. Slides available at: chapel-lang.org/ChapelCon/2024/kuno.pdf
ChapelCon '24: Chplx: an HPX Foundation for Chapel
Просмотров 27Месяц назад
This is Chris Taylor's recorded talk for ChapelCon'24. Slides available at: chapel-lang.org/ChapelCon/2024/taylor.pdf
ChapelCon '24: Chapel in a Petabyte-Scale GPU Database Engine with Voltron Data’s Theseus
Просмотров 37Месяц назад
This is Fernanda Foerrter and Trent Nelson's ChapelCon'24 talk, recorded live on June 7, 2024. Slides available at: chapel-lang.org/ChapelCon/2024/nelson.pdf
ChapelCon '24: Arrays as Arguments in First-Class Functions-the Levenberg-Marquardt Algorithm
Просмотров 37Месяц назад
This is Nelson Dias's ChapelCon'24 talk, recorded live on June 7, 2024. Please note that the full title of the talk is "Arrays as Arguments in First-Class Functions-the Levenberg-Marquardt Algorithm in Chapel" (which went over RUclips's title character count limit). Slides available at: chapel-lang.org/ChapelCon/2024/dias.pdf
ChapelCon '24: Implementing Imaginary Elementary Mathematical Functions
Просмотров 8Месяц назад
This is Damian McGuckin's pre-recorded ChapelCon'24 talk with Q&A recorded live on June 7, 2024. Slides available at: chapel-lang.org/ChapelCon/2024/mcguckin.pdf
ChapelCon '24: On the Design of Graph Analytical Software in Chapel
Просмотров 25Месяц назад
This is Oliver Alvarado Rodriguez's ChapelCon'24 talk, recorded live on June 7, 2024. Slides available at: chapel-lang.org/ChapelCon/2024/rodriguez.pdf
ChapelCon '24: Unbalanced Tree-Search at Scale Using the Chapel's DistributedBag Module
Просмотров 19Месяц назад
ChapelCon '24: Unbalanced Tree-Search at Scale Using the Chapel's DistributedBag Module
ChapelCon '24: Exploring Machine Learning Capabilities in Chapel: An Internship Journey
Просмотров 38Месяц назад
ChapelCon '24: Exploring Machine Learning Capabilities in Chapel: An Internship Journey
ChapelCon '24: Building a Chapel Curriculum on Exercism
Просмотров 31Месяц назад
ChapelCon '24: Building a Chapel Curriculum on Exercism
ChapelCon '24 Keynote: A Case for Parallel-First Languages in a Post-Serial, Accelerated World
Просмотров 286Месяц назад
ChapelCon '24 Keynote: A Case for Parallel-First Languages in a Post-Serial, Accelerated World
ChapelCon '24: Braiding a Million Threads: Scalable GPU Sort on Frontier
Просмотров 26Месяц назад
ChapelCon '24: Braiding a Million Threads: Scalable GPU Sort on Frontier
ChapelCon '24: Investigating Portability for Tree-based Optimization on GPU-powered Clusters
Просмотров 11Месяц назад
ChapelCon '24: Investigating Portability for Tree-based Optimization on GPU-powered Clusters
ChapellCon '24: Performance Portability of Chapel on Diverse Architectures
Просмотров 33Месяц назад
ChapellCon '24: Performance Portability of Chapel on Diverse Architectures
ChapelCon '24: The Computer Language Benchmarks Game and Chapel 2.0
Просмотров 60Месяц назад
ChapelCon '24: The Computer Language Benchmarks Game and Chapel 2.0
ChapelCon '24: Advanced Editor Tooling for Chapel
Просмотров 54Месяц назад
ChapelCon '24: Advanced Editor Tooling for Chapel
ChapelCon '24: WWU Chapel Debugger
Просмотров 23Месяц назад
ChapelCon '24: WWU Chapel Debugger
ChapelCon '24: State of the Chapel Project
Просмотров 64Месяц назад
ChapelCon '24: State of the Chapel Project
ChapelCon '24: Welcome to ChapelCon'24
Просмотров 225Месяц назад
ChapelCon '24: Welcome to ChapelCon'24
GitHub Codespaces
Просмотров 663 месяца назад
GitHub Codespaces
CHIUW 2023 Keynote: PGAS Programming Models: My 20-year Perspective
Просмотров 220Год назад
CHIUW 2023 Keynote: PGAS Programming Models: My 20-year Perspective
CHIUW2023: Towards a Scalable Load Balancing for Productivity-Aware Tree-Search
Просмотров 61Год назад
CHIUW2023: Towards a Scalable Load Balancing for Productivity-Aware Tree-Search

Комментарии

  • @edbertkwesi4931
    @edbertkwesi4931 7 дней назад

    this language is very powerful

  • @AozenDreyar
    @AozenDreyar 19 дней назад

    does chapel have any jobs?

    • @ChapelLanguage
      @ChapelLanguage 7 дней назад

      @AozenDreyar: Thanks for your inquiry. Unfortunately, we don't have any open positions at HPE today (or elsewhere in the Chapel community that I'm aware of). As opportunities become available, they'll be advertised on the Chapel jobs page (chapel-lang.org/jobs.html ), the HPE careers page (careers.hpe.com/us/en ), and our social media channels (chapel-lang.org/community.html ).

  • @nemea6698
    @nemea6698 20 дней назад

    Why was it called chapel?

    • @ChapelLanguage
      @ChapelLanguage 7 дней назад

      @neme6698: Thanks for the question! Chapel was developed as part of a larger Cray project named Cascade, and its name came from "Cascade High-Productivity Language" or CHPL, which evolved into Chapel. For more information about Chapel's origins, see chapel-lang.org/publications/PMfPC-Chapel.pdf

    • @nemea6698
      @nemea6698 7 дней назад

      @@ChapelLanguage is the name affected by christianity? I am a christian

    • @ChapelLanguage
      @ChapelLanguage 6 дней назад

      @@nemea6698: It is not.

    • @nemea6698
      @nemea6698 6 дней назад

      @@ChapelLanguage okay

    • @nemea6698
      @nemea6698 6 дней назад

      @@ChapelLanguage okay

  • @eduardabramovich1216
    @eduardabramovich1216 Месяц назад

    I wish you created a comprehensive series of tutorials showcasing chapel through examples.

    • @ChapelLanguage
      @ChapelLanguage 7 дней назад

      @eduardabramovich1216: Thanks for your interest! We're actually working on creating more material along the lines you suggest. For example, we've just published a playlist of short demos teaching aspects of programming GPUs using Chapel: ruclips.net/p/PLuqM5RJ2KYFgNM62w9QdRTcWPpf3rfSwr We plan to do more short-form videos like this for traditional CPU executions in Chapel as well.

  • @sB3rg
    @sB3rg Месяц назад

    Something wrong with your audio =/

    • @ChapelLanguage
      @ChapelLanguage 7 дней назад

      @sB3rg: Sounds OK from here in an incognito tab. What are you experiencing on your end?

  • @edbertkwesi4931
    @edbertkwesi4931 Месяц назад

    this this is very powerful i once tried it

  • @encapsulatio
    @encapsulatio Месяц назад

    One thing to take into consideration is that at the moment the best code generator for different programming languages is Clause Sonnet 3.5 which is freely available to use by people. Yes it's better than ChatGpt 4o at writing code. It could be customized to always have a up to date retention of all the documentation of Chapel 2.0 and use it as a tutor for beginner programmers who have not programmed before ...like me. The only thing Chapel team needs to make sure before making this available to people is that all the programming concepts unique to Chapel are actually properly understood and explained by Claude Sonnet 3.5. If that is possible anyone can just use one of their favorite beginner programming books made for complete beginners in some popular language, paste a whole chapter and make Claude rewrite that chapter, but rewrite it in such a way that the exercises and structure of exercises remains identical but Clause only rewrites the syntax and adds explanations of the semantics of Chapel for every exercise so it is truly idiomatic Chapel. This way you don't have to invest in making extensive beginner friendly resources, just have to invest time in making sure Claude understands all the syntax and semantics and in depth all the programming model from Chapel. People are using max 3-4 well designed prompts for building whole very specialized apps completely in Claude. There is no reason to believe why Claude can not be used successfully as the personalized tutor for everyone wanting to learn Chapel, whether a complete beginner who has not seen or used a programming language before or an intermediate programmer in Python that wants to make sure he understands completely all the nuances of Chapel's parallel programming model.

    • @ChapelLanguage
      @ChapelLanguage Месяц назад

      This discussion would probably get more eyes on it if it were to take place on Chapel's Discourse or GitHub issues. See chapel-lang.org/community.html for links if needed.

  • @encapsulatio
    @encapsulatio Месяц назад

    The link to slides is broken: The requested URL was not found on this server. Additionally, a 404 Not Found error was encountered while trying to use an ErrorDocument to handle the request.

    • @ChapelLanguage
      @ChapelLanguage Месяц назад

      Thanks for pointing that out @encapsulatio, it should be fixed now! (we had to change formats for the slides for this talk, but forgot to update this link).

    • @encapsulatio
      @encapsulatio Месяц назад

      @@ChapelLanguage What about what I wrote in my other reply to you on this same video?

    • @ChapelLanguage
      @ChapelLanguage Месяц назад

      ​@@encapsulatio : RUclips wasn't making it particularly visible to us, but searching for it now, we see it and will pass it along to others.

  • @tanvirhossain18
    @tanvirhossain18 Год назад

    Hi what is the meaning of -- if ( e is local) ?

    • @ChapelLanguage
      @ChapelLanguage 3 месяца назад

      @tanvirhoussain: In this context (distributed memory computing), I believe the intention is "if e is local to the locale (compute node) on which the current task is running"

  • @LadyTink
    @LadyTink Год назад

    Hey so i've been messing around with chapel for a few days. One thing I find annoying, is the way random numbers are handled in forall vs for loops. Like I was testing performance differences and found that the normal mechanism for random numbers resulted in a huge speed loss. Instead I need to use var parallelUnsafe = new RandomStream(real, parSafe=false); which is clunky, when if the random stream is being set inside for loop, outside determinism, I honestly feel the default would make more sense to default one way in tasks like this. I might not be the target audience for this language. However I was looking for a language that was like erlang with how wide it can scale, while having great performance, which chapel seems to have. This is just a minor usability/convenience aspect

  • @user-de8rd4yn6j
    @user-de8rd4yn6j Год назад

    🕊✝ ВАРИАНТЫ РАЗРУШЕНИЯ МИРА ПЕРЕД "ВЕЛИКОЙ СКОРБЬЮ"(ПОСЛЕДНИЕ ВРЕМЕНА). ruclips.net/video/j-EYHyL47-8/видео.html

  • @AJ-et3vf
    @AJ-et3vf 2 года назад

    Great video sir. Thank you

  • @tommacdonald8184
    @tommacdonald8184 4 года назад

    Nice presentation about a useful feature.