http://CppCon.org
—
Presentation Slides, PDFs, Source Code and other presenter materials are available at: https://github.com/cppcon/cppcon2016
—
SIMD extensions have been a feature of choice for processor manufacturers for a couple of decades. Designed to exploit data parallelism in applications at the instruction level, these extensions still require a high level of expertise or the use of potentially fragile compiler support or vendor-specific libraries. While a large fraction of their theoretical accelerations can be obtained using such tools, exploiting such hardware becomes tedious as soon as application portability across hardware is required.
Accessing such capabilities directly from C++ code could be a major improvements in a lot of use cases. Different take on this has been proposed either by the community or as an actual standard proposal. Solutions include pragma based annotations, standard algorithms policies, full blown compiler support and libraries.
In this talk we will present one such solution - the Boost.SIMD library (currently being proposed as such) which takes a library approach to this issues.
We will go over the basic notion required to grasp SIMD programming in general. Then, we'll discuss the different existing approaches. We will describe Boost.SIMD API and API design to demonstrate how it solves issues raised by the actual idiomatic way of writting SIMD enabled code. Design issues like standard algorithm integration, memory handling and how to fill the gaps in SIMD instructions sets will be discussed. Finally, we show its performances with respect to a subset of well known benchmarks.
—
Joel Falcou
CTO, NUMSCALE
Joel Falcou is NumScale CTO. NumScale mission is to assist businesses in the exploration and subsequently the mastery of high-performance computing systems. | | He is also an assistant professor at the University Paris-Sud and researcher at the Laboratoire de Recherche d’Informatique in Orsay, France.
—
Videos Filmed & Edited by Bash Films: http://www.BashFilms.com
—
Presentation Slides, PDFs, Source Code and other presenter materials are available at: https://github.com/cppcon/cppcon2016
—
SIMD extensions have been a feature of choice for processor manufacturers for a couple of decades. Designed to exploit data parallelism in applications at the instruction level, these extensions still require a high level of expertise or the use of potentially fragile compiler support or vendor-specific libraries. While a large fraction of their theoretical accelerations can be obtained using such tools, exploiting such hardware becomes tedious as soon as application portability across hardware is required.
Accessing such capabilities directly from C++ code could be a major improvements in a lot of use cases. Different take on this has been proposed either by the community or as an actual standard proposal. Solutions include pragma based annotations, standard algorithms policies, full blown compiler support and libraries.
In this talk we will present one such solution - the Boost.SIMD library (currently being proposed as such) which takes a library approach to this issues.
We will go over the basic notion required to grasp SIMD programming in general. Then, we'll discuss the different existing approaches. We will describe Boost.SIMD API and API design to demonstrate how it solves issues raised by the actual idiomatic way of writting SIMD enabled code. Design issues like standard algorithm integration, memory handling and how to fill the gaps in SIMD instructions sets will be discussed. Finally, we show its performances with respect to a subset of well known benchmarks.
—
Joel Falcou
CTO, NUMSCALE
Joel Falcou is NumScale CTO. NumScale mission is to assist businesses in the exploration and subsequently the mastery of high-performance computing systems. | | He is also an assistant professor at the University Paris-Sud and researcher at the Laboratoire de Recherche d’Informatique in Orsay, France.
—
Videos Filmed & Edited by Bash Films: http://www.BashFilms.com
c thomas howell CppCon 2016: Joel Falcou “Designing C++ portable SIMD support" | |
33 Likes | 33 Dislikes |
3,859 views views | 63.8K followers |
Education | Upload TimePublished on 7 Oct 2016 |
Related keywords
c train schedule,c diff,c corporation,c town,c train service,video services uiuc,video services unlimited,c section scar,c to k,video services inc,video services contract,c battery,video services of america,c span,video services on switch,video services list,c street,c suite,video services group,video services of va,video services near me,video services like netflix,c chord guitar,video services boise,c train status,video services conroe tx,video services forum,c section,video services other than youtube,video services group inc,c train,video services like youtube,c to f,c stock,c reactive protein,video services before youtube,video services pricing,video services contract template,c elegans,
Không có nhận xét nào:
Đăng nhận xét