Тёмный
No video :(

C++ Standard Parallelism - Bryce Adelstein Lelbach - CppCon 2021 

CppCon
Подписаться 152 тыс.
Просмотров 13 тыс.
50% 1

cppcon.org/
github.com/Cpp...
---
Imagine writing parallel code that can run on any platform - CPUs, GPUs, DPUs, specialized accelerators, etc. It's no longer just a dream - you can do it in Standard C++!
Parallelism is increasingly common in software, from supercomputer simulations to mobile applications. But writing parallel code is increasingly challenging, due to an explosion of diversity in hardware, a trend that's likely to continue into the future. To meet this challenge, the Standard C++ Committee has developed a roadmap for C++ Standard Parallelism, a parallel programming model that will be portable to all platforms while delivering reasonable performance and efficiency for most use cases.
Our vision of C++ Standard Parallelism consists of three key components:
* Common parallel algorithms that dispatch to vendor-optimized parallel libraries
* Tools to write your own parallel algorithms that run anywhere.
* Mechanisms for composing parallel invocations of algorithms into task graphs.
In this talk, we'll dive into this roadmap - we'll discuss what we already have that you can use today, what's coming down the line, and where the future may lead us.
---
Bryce Lelbach
Bryce Adelstein Lelbach has spent over a decade developing software libraries and programming languages. He is the HPC Programming Models Architect at NVIDIA, where he leads programming language standardization efforts and drives the technical roadmap for NVIDIA's HPC compilers and libraries. Bryce is passionate about C++ and is one of the leaders of the C++ community. He is the
chair of INCITS/PL22, the US standards committee for programming languages and the Standard C++ Library Evolution group. Bryce is the program chair for the C++Now and CppCon conferences, and the chief organizer of the Bay Area C++ User Group. On the C++ Committee, he has personally worked on the C++17 parallel algorithms, executors, futures, senders/receivers, multidimensional arrays, and modules. He is one of the initial developers of the HPX parallel runtime system. He also helped start the LLVMLinux initiative and has occasionally contributed to the Boost C++ libraries.
---
Videos Streamed & Edited by Digital Medium: online.digital-...
*--*

Опубликовано:

 

27 авг 2024

Поделиться:

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист
Посмотреть позже
Комментарии : 15   
@mmocny
@mmocny 2 года назад
The quality of this presentation is outstanding. Thank you Bryce.
@VityokOrgUa
@VityokOrgUa 2 года назад
wonderful talk, very accessible presentation of a rather complex subject
@KhalilEstell
@KhalilEstell 2 года назад
Great presentation Bryce! A lot of content to get through but I think you did a great job making it reasonable to understand. Cheers!
@bearwolffish
@bearwolffish Год назад
just keeps getting better
@funkybibimbap6972
@funkybibimbap6972 2 года назад
I might have missed that while watching the video, but if you're wondering like me what that stdv:: namespace is, it's actually std::ranges::views::
@VoidloniXaarii
@VoidloniXaarii 11 месяцев назад
This was mind-blowing... Thank you
@jeff4615
@jeff4615 6 месяцев назад
After watching this video, I finally knew that I don’t know C++ at all⋯
@thebasicmaterialsproject1892
@thebasicmaterialsproject1892 2 года назад
maybe a little too advanced for me this one great lecture
@Chris-yw5is
@Chris-yw5is 2 года назад
9:25 real shame that the implementation can lie to you like that. Ran into that earlier today and that caused loads of headaches. Hope the coming things are gonna be more reliable. Im definitely excited for mdspan tho. Great talk really appreciate the graphics and syntax highlighting.
@peterevans6752
@peterevans6752 2 года назад
Great Presentation! mdspan and execution look like high utility features. I hope most of makes it into c++23. I think it did, but I'm just learning the papers and how to track them.
@bryce.ferenczi
@bryce.ferenczi 2 года назад
mdspan will be great, I've recently had to transform a large image into a batch of chips with a dimension permutation and the only thing that jumped out at me without finding and including another library was a triple nested for loop with index calculation.
@embeddor2230
@embeddor2230 2 года назад
man std::transform really rocks :D
@victoreijkhout6146
@victoreijkhout6146 2 года назад
So md-span will still have each dimension starting at zero? Maybe C++26 will catch up with Fortran90 in having a declarable lower bound.
@sinamirmahmoud7606
@sinamirmahmoud7606 2 года назад
wow
@giuliograziani3261
@giuliograziani3261 2 года назад
when does nvidia plan to release it for windows?
Далее
Conquering C++20 Ranges - Tristan Brindle - CppCon 2021
1:06:07
Whoa
01:00
Просмотров 33 млн
А ВЫ УМЕЕТЕ ПЛАВАТЬ?? #shorts
00:21
Просмотров 1,5 млн
СЕРЕГА ПИРАТ - TEAM SPIRIT
02:37
Просмотров 302 тыс.
C++ Super Optimization: 1000X Faster
15:33
Просмотров 314 тыс.
Back to Basics: Concurrency - Mike Shah - CppCon 2021
1:02:07
The Art of Code - Dylan Beattie
1:00:49
Просмотров 4,7 млн