A GPU Spatial Processing System for CHIME
© 2020 World Scientific Publishing Company. We present an overview of the Graphics Processing Unit (GPU)-based spatial processing system created for the Canadian Hydrogen Intensity Mapping Experiment (CHIME). The design employs AMD S9300x2 GPUs and readily available commercial hardware in its proces...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
World Scientific Pub Co Pte Lt
2021
|
| Online Access: | https://hdl.handle.net/1721.1/132478 |
| _version_ | 1811083738467008512 |
|---|---|
| author | Denman, N Renard, A Vanderlinde, K Berger, P Masui, K Tretyakov, I |
| author_facet | Denman, N Renard, A Vanderlinde, K Berger, P Masui, K Tretyakov, I |
| author_sort | Denman, N |
| collection | MIT |
| description | © 2020 World Scientific Publishing Company. We present an overview of the Graphics Processing Unit (GPU)-based spatial processing system created for the Canadian Hydrogen Intensity Mapping Experiment (CHIME). The design employs AMD S9300x2 GPUs and readily available commercial hardware in its processing nodes to provide a cost- and power-efficient processing substrate. These nodes are supported by a liquid-cooling system which allows continuous operation with modest power consumption and in all but the most adverse conditions. Capable of continuously correlating 2048 receiver-polarizations across 400MHz of bandwidth, the CHIME X-engine constitutes the most powerful radio correlator currently in existence. It receives 6.6Tb/s of channelized data from CHIME's FPGA-based F-engine, and the primary correlation task requires 8.39×1014 complex multiply-and-accumulate operations per second. The same system also provides formed-beam data products to commensal FRB and Pulsar experiments; it constitutes a general spatial-processing system of unprecedented scale and capability, with correspondingly great challenges in computation, data transport, heat dissipation, and interference shielding. |
| first_indexed | 2024-09-23T12:38:16Z |
| format | Article |
| id | mit-1721.1/132478 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T12:38:16Z |
| publishDate | 2021 |
| publisher | World Scientific Pub Co Pte Lt |
| record_format | dspace |
| spelling | mit-1721.1/1324782021-09-21T03:35:02Z A GPU Spatial Processing System for CHIME Denman, N Renard, A Vanderlinde, K Berger, P Masui, K Tretyakov, I © 2020 World Scientific Publishing Company. We present an overview of the Graphics Processing Unit (GPU)-based spatial processing system created for the Canadian Hydrogen Intensity Mapping Experiment (CHIME). The design employs AMD S9300x2 GPUs and readily available commercial hardware in its processing nodes to provide a cost- and power-efficient processing substrate. These nodes are supported by a liquid-cooling system which allows continuous operation with modest power consumption and in all but the most adverse conditions. Capable of continuously correlating 2048 receiver-polarizations across 400MHz of bandwidth, the CHIME X-engine constitutes the most powerful radio correlator currently in existence. It receives 6.6Tb/s of channelized data from CHIME's FPGA-based F-engine, and the primary correlation task requires 8.39×1014 complex multiply-and-accumulate operations per second. The same system also provides formed-beam data products to commensal FRB and Pulsar experiments; it constitutes a general spatial-processing system of unprecedented scale and capability, with correspondingly great challenges in computation, data transport, heat dissipation, and interference shielding. 2021-09-20T18:22:38Z 2021-09-20T18:22:38Z 2020-11-03T16:11:35Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/132478 en 10.1142/S2251171720500142 Journal of Astronomical Instrumentation Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf World Scientific Pub Co Pte Lt arXiv |
| spellingShingle | Denman, N Renard, A Vanderlinde, K Berger, P Masui, K Tretyakov, I A GPU Spatial Processing System for CHIME |
| title | A GPU Spatial Processing System for CHIME |
| title_full | A GPU Spatial Processing System for CHIME |
| title_fullStr | A GPU Spatial Processing System for CHIME |
| title_full_unstemmed | A GPU Spatial Processing System for CHIME |
| title_short | A GPU Spatial Processing System for CHIME |
| title_sort | gpu spatial processing system for chime |
| url | https://hdl.handle.net/1721.1/132478 |
| work_keys_str_mv | AT denmann agpuspatialprocessingsystemforchime AT renarda agpuspatialprocessingsystemforchime AT vanderlindek agpuspatialprocessingsystemforchime AT bergerp agpuspatialprocessingsystemforchime AT masuik agpuspatialprocessingsystemforchime AT tretyakovi agpuspatialprocessingsystemforchime AT denmann gpuspatialprocessingsystemforchime AT renarda gpuspatialprocessingsystemforchime AT vanderlindek gpuspatialprocessingsystemforchime AT bergerp gpuspatialprocessingsystemforchime AT masuik gpuspatialprocessingsystemforchime AT tretyakovi gpuspatialprocessingsystemforchime |