$\texttt{cunuSHT}$: GPU Accelerated Spherical Harmonic Transforms on Arbitrary Pixelizations
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914843344240640 |
|---|---|
| author | Belkner, Sebastian Duivenvoorden, Adriaan J. Carron, Julien Schaeffer, Nathanael Reinecke, Martin |
| author_facet | Belkner, Sebastian Duivenvoorden, Adriaan J. Carron, Julien Schaeffer, Nathanael Reinecke, Martin |
| contents | We present $\texttt{cunusht}$, a general-purpose Python package that wraps a highly efficient CUDA implementation of the nonuniform spin-$0$ spherical harmonic transform. The method is applicable to arbitrary pixelization schemes, including schemes constructed from equally-spaced iso-latitude rings as well as completely nonuniform ones. The algorithm has an asymptotic scaling of $\mathrm{O}{(\ell_{\rm max}^3)}$ for maximum multipole $\ell_{\rm max}$ and achieves machine precision accuracy. While $\texttt{cunusht}$ is developed for applications in cosmology in mind, it is applicable to various other interpolation problems on the sphere. We outperform the fastest available CPU algorithm by a factor of up to 5 for problems with a nonuniform pixelization and $\ell_{\rm max}>4\cdot10^3$ when comparing a single modern GPU to a modern 32-core CPU. This performance is achieved by utilizing the double Fourier sphere method in combination with the nonuniform fast Fourier transform and by avoiding transfers between the host and device. For scenarios without GPU availability, $\texttt{cunusht}$ wraps existing CPU libraries. $\texttt{cunusht}$ is publicly available and includes tests, documentation, and demonstrations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_14542 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | $\texttt{cunuSHT}$: GPU Accelerated Spherical Harmonic Transforms on Arbitrary Pixelizations Belkner, Sebastian Duivenvoorden, Adriaan J. Carron, Julien Schaeffer, Nathanael Reinecke, Martin Instrumentation and Methods for Astrophysics Cosmology and Nongalactic Astrophysics We present $\texttt{cunusht}$, a general-purpose Python package that wraps a highly efficient CUDA implementation of the nonuniform spin-$0$ spherical harmonic transform. The method is applicable to arbitrary pixelization schemes, including schemes constructed from equally-spaced iso-latitude rings as well as completely nonuniform ones. The algorithm has an asymptotic scaling of $\mathrm{O}{(\ell_{\rm max}^3)}$ for maximum multipole $\ell_{\rm max}$ and achieves machine precision accuracy. While $\texttt{cunusht}$ is developed for applications in cosmology in mind, it is applicable to various other interpolation problems on the sphere. We outperform the fastest available CPU algorithm by a factor of up to 5 for problems with a nonuniform pixelization and $\ell_{\rm max}>4\cdot10^3$ when comparing a single modern GPU to a modern 32-core CPU. This performance is achieved by utilizing the double Fourier sphere method in combination with the nonuniform fast Fourier transform and by avoiding transfers between the host and device. For scenarios without GPU availability, $\texttt{cunusht}$ wraps existing CPU libraries. $\texttt{cunusht}$ is publicly available and includes tests, documentation, and demonstrations. |
| title | $\texttt{cunuSHT}$: GPU Accelerated Spherical Harmonic Transforms on Arbitrary Pixelizations |
| topic | Instrumentation and Methods for Astrophysics Cosmology and Nongalactic Astrophysics |
| url | https://arxiv.org/abs/2406.14542 |