Exploiting the chaotic behaviour of atmospheric models with reconfigurable architectures
Reconfigurable architectures are becoming mainstream: Amazon, Microsoft and IBM are supporting such architectures in their data centres. The computationally intensive nature of atmospheric modelling is an attractive target for hardware acceleration using reconfigurable computing. Performance of hard...
| Autores principales: | , , , , |
|---|---|
| Formato: | Journal article |
| Lenguaje: | English |
| Publicado: |
Elsevier
2017
|
| _version_ | 1826293008596729856 |
|---|---|
| author | Russell, F Düben, P Niu, X Luk, W Palmer, T |
| author_facet | Russell, F Düben, P Niu, X Luk, W Palmer, T |
| author_sort | Russell, F |
| collection | OXFORD |
| description | Reconfigurable architectures are becoming mainstream: Amazon, Microsoft and IBM are supporting such architectures in their data centres. The computationally intensive nature of atmospheric modelling is an attractive target for hardware acceleration using reconfigurable computing. Performance of hardware designs can be improved through the use of reduced-precision arithmetic, but maintaining appropriate accuracy is essential. We explore reduced-precision optimisation for simulating chaotic systems, targeting atmospheric modelling, in which even minor changes in arithmetic behaviour will cause simulations to diverge quickly. The possibility of equally valid simulations having differing outcomes means that standard techniques for comparing numerical accuracy are inappropriate. We use the Hellinger distance to compare statistical behaviour between reduced-precision CPU implementations to guide reconfigurable designs of a chaotic system, then analyse accuracy, performance and power efficiency of the resulting implementations. Our results show that with only a limited loss in accuracy corresponding to less than 10% uncertainty in input parameters, the throughput and energy efficiency of a single-precision chaotic system implemented on a Xilinx Virtex-6 SX475T Field Programmable Gate Array (FPGA) can be more than doubled. |
| first_indexed | 2024-03-07T03:23:28Z |
| format | Journal article |
| id | oxford-uuid:b83b2f2a-aca2-4ba0-a1f4-f743b993245c |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T03:23:28Z |
| publishDate | 2017 |
| publisher | Elsevier |
| record_format | dspace |
| spelling | oxford-uuid:b83b2f2a-aca2-4ba0-a1f4-f743b993245c2022-03-27T04:54:35ZExploiting the chaotic behaviour of atmospheric models with reconfigurable architecturesJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:b83b2f2a-aca2-4ba0-a1f4-f743b993245cEnglishSymplectic Elements at OxfordElsevier2017Russell, FDüben, PNiu, XLuk, WPalmer, TReconfigurable architectures are becoming mainstream: Amazon, Microsoft and IBM are supporting such architectures in their data centres. The computationally intensive nature of atmospheric modelling is an attractive target for hardware acceleration using reconfigurable computing. Performance of hardware designs can be improved through the use of reduced-precision arithmetic, but maintaining appropriate accuracy is essential. We explore reduced-precision optimisation for simulating chaotic systems, targeting atmospheric modelling, in which even minor changes in arithmetic behaviour will cause simulations to diverge quickly. The possibility of equally valid simulations having differing outcomes means that standard techniques for comparing numerical accuracy are inappropriate. We use the Hellinger distance to compare statistical behaviour between reduced-precision CPU implementations to guide reconfigurable designs of a chaotic system, then analyse accuracy, performance and power efficiency of the resulting implementations. Our results show that with only a limited loss in accuracy corresponding to less than 10% uncertainty in input parameters, the throughput and energy efficiency of a single-precision chaotic system implemented on a Xilinx Virtex-6 SX475T Field Programmable Gate Array (FPGA) can be more than doubled. |
| spellingShingle | Russell, F Düben, P Niu, X Luk, W Palmer, T Exploiting the chaotic behaviour of atmospheric models with reconfigurable architectures |
| title | Exploiting the chaotic behaviour of atmospheric models with reconfigurable architectures |
| title_full | Exploiting the chaotic behaviour of atmospheric models with reconfigurable architectures |
| title_fullStr | Exploiting the chaotic behaviour of atmospheric models with reconfigurable architectures |
| title_full_unstemmed | Exploiting the chaotic behaviour of atmospheric models with reconfigurable architectures |
| title_short | Exploiting the chaotic behaviour of atmospheric models with reconfigurable architectures |
| title_sort | exploiting the chaotic behaviour of atmospheric models with reconfigurable architectures |
| work_keys_str_mv | AT russellf exploitingthechaoticbehaviourofatmosphericmodelswithreconfigurablearchitectures AT dubenp exploitingthechaoticbehaviourofatmosphericmodelswithreconfigurablearchitectures AT niux exploitingthechaoticbehaviourofatmosphericmodelswithreconfigurablearchitectures AT lukw exploitingthechaoticbehaviourofatmosphericmodelswithreconfigurablearchitectures AT palmert exploitingthechaoticbehaviourofatmosphericmodelswithreconfigurablearchitectures |