Celerity: High-Level C++ for Accelerator Clusters

Celerity: High-Level C++ for Accelerator Clusters
复制标题

Celerity:加速器集群的高级 C

DOI:
10.1007/978-3-030-29400-7_21
复制
发表时间:
2019
期刊:
影响因子:
--
通讯作者:
Thomas Fahringer
Thomas Fahringer
中科院分区:
--
文献类型:
--
作者:
Peter Thoman;Philip Salzmann;Biagio Cosenza;Thomas Fahringer

文献摘要

参考文献

被引文献

相似文献

面对 CPU 单线程性能增长不断放缓的情况,科学和工程界越来越多地转向加速器并行化来应对不断增长的应用程序工作负载。针对分布式内存加速器集群的现有方法带来了严重的可编程性障碍和维护负担。Celerity 编程环境旨在使开发人员能够相对轻松地将 C++ 应用程序扩展到加速器集群,同时利用和扩展 SYCL 特定于域的嵌入式语言。通过让用户提供有关如何在计算内核中访问数据的最少信息,Celerity 自动分配工作和数据。我们引入了 Celerity C++ API 以及原型实现,证明现有 SYCL 代码只需遵循既定习惯用法的一小部分更改即可引入分布式内存集群。 Celerity 原型运行时实现与更传统的分布式内存加速器编程方法(例如 MPI+OpenCL)具有相当的性能,并且实现复杂性显着降低。
In the face of ever-slowing single-thread performance growth for CPUs, the scientific and engineering communities increasingly turn to accelerator parallelization to tackle growing application workloads. Existing means of targeting distributed memory accelerator clusters impose severe programmability barriers and maintenance burdens.The Celerity programming environment seeks to enable developers to scale C++ applications to accelerator clusters with relative ease, while leveraging and extending the SYCL domain-specific embedded language. By having users provide minimal information about how data is accessed within compute kernels, Celerity automatically distributes work and data.We introduce the Celerity C++ API as well as a prototype implementation, demonstrating that existing SYCL code can be brought to distributed memory clusters with only a small set of changes that follow established idioms. The Celerity prototype runtime implementation is shown to have comparable performance to more traditional approaches to distributed memory accelerator programming, such as MPI+OpenCL, with significantly lower implementation complexity.
使用基于顺序任务的编程模型在超级计算机上实现高性能
DOI: 10.1109/tpds.2017.2766064
发表时间: 2017
影响因子: 5.3
作者:
E. Agullo;Olivier Aumage;Mathieu Faverge;N. Furmento;Florent Pruvost;Marc Sergent;Samuel Thibault
通讯作者: Samuel Thibault
DOI: 10.1016/j.jss.2016.02.002
发表时间: 2017
期刊: J. Syst. Softw.
影响因子: --
作者:
A. Meade;D. Deeptimahanti;J. Buckley;J. Collins
通讯作者: J. Collins
DOI: 10.1109/hpcs.2017.109
发表时间: 2017
期刊: 2017 International Conference on High Performance Computing & Simulation (HPCS)
影响因子: --
作者:
Biagio Peccerillo;S. Bartolini
通讯作者: S. Bartolini
DOI: 10.1007/s11227-018-2238-4
发表时间: 2018-04-01
影响因子: 3.3
作者:
Thoman, Peter;Dichev, Kiril;Nikolopoulos, Dimitrios S.
通讯作者: Nikolopoulos, Dimitrios S.