A GPU-based Correlator X-engine Implemented on the CHIME Pathfinder

被引：0

作者：

Denman, Nolan ^{[1
,2
]}

Amiri, Mandana ^{[3
]}

Bandura, Kevin ^{[4
]}

Connor, Liam ^{[1
,2
,5
]}

Dobbs, Matt ^{[4
,6
]}

Fandino, Mateus ^{[3
]}

Halpern, Mark ^{[3
]}

Hincks, Adam ^{[3
]}

Hinshaw, Gary ^{[3
]}

Hofer, Carolin ^{[3
]}

Klages, Peter ^{[1
]}

Masui, Kiyoshi ^{[3
,6
]}

Parra, Juan Mena ^{[4
]}

Newburgh, Laura ^{[1
]}

Recnik, Andre ^{[1
]}

Shaw, J. Richard ^{[5
]}

Sigurdson, Kris ^{[3
]}

Smith, Kendrick ^{[7
]}

Vanderlinde, Keith ^{[1
,2
]}

机构：

[1] Univ Toronto, Dunlap Inst, Toronto, ON M5S 1A1, Canada

[2] Univ Toronto, Dept Astron & Astrophys, Toronto, ON M5S 1A1, Canada

[3] Univ British Columbia, Dept Phys & Astron, Vancouver, BC V5Z 1M9, Canada

[4] McGill Univ, Dept Phys, Montreal, PQ H3A 2T5, Canada

[5] Canadian Inst Theoret Astrophys, Toronto, ON, Canada

[6] Canadian Inst Adv Res, CIFAR Program Cosmol & Grav, Toronto, ON, Canada

[7] Perimeter Inst Theoret Phys, Waterloo, ON, Canada

来源：

PROCEEDINGS OF THE ASAP2015 2015 IEEE 26TH INTERNATIONAL CONFERENCE ON APPLICATION-SPECIFIC SYSTEMS, ARCHITECTURES AND PROCESSORS | 2015年

关键词：

D O I：

暂无

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

We present the design and implementation of a custom GPU-based compute cluster that provides the correlation X-engine of the CHIME Pathfinder radio telescope. It is among the largest such systems in operation, correlating 32,896 baselines (256 inputs) over 400MHz of radio bandwidth. Making heavy use of consumer-grade parts and a custom software stack, the system was developed at a small fraction of the cost of comparable installations. Unlike existing GPU backends, this system is built around OpenCL kernels running on consumer-level AMD GPUs, taking advantage of low-cost hardware and leveraging packed integer operations to double algorithmic efficiency. The system achieves the required 105 TOPS in a 10kW power envelope, making it one of the most power-efficient X-engines in use today.

引用

页码：35 / 40

页数：6

共 43 条

[41] Technical note: A GPU-based shared Monte Carlo method for fast photon transport in multi-energy x-ray exposures
Zhou, Yiwen
Deng, Wenxin
Kang, Jing
Xia, Jinqiu
Yang, Yingjie
Li, Bin
Zhang, Yuqin
Qi, Hongliang
Wu, WangJiang
Qi, Mengke
Zhou, Linghong
Ma, Jianhui
Xu, Yuan
MEDICAL PHYSICS, 2024, 51 (11) : 8390 - 8398
[42] A GPU-based multi-resolution approach to iterative reconstruction algorithms in x-ray 3D dual spectral computed tomography
Hu, Jingjing
Zhao, Xing
Zhang, Huitao
NEUROCOMPUTING, 2016, 215 : 71 - 81
[43] Virtual Clinical Trials in 2D and 3D X-ray Breast Imaging and Dosimetry: Comparison of CPU-Based and GPU-Based Monte Carlo Codes
Mettivier, Giovanni
Sarno, Antonio
Lai, Youfang
Golosio, Bruno
Fanti, Viviana
Italiano, Maria Elena
Jia, Xun
Russo, Paolo
CANCERS, 2022, 14 (04)

← 1 2 3 4 5 →