SciPost Submission Page
A Python GPU-accelerated solver for the Gross-Pitaevskii equation and applications to many-body cavity QED
by Lorenzo Fioroni, Luca Gravina, Justyna Stefaniak, Alexander Baumgärtner, Fabian Finger, Davide Dreon, Tobias Donner
This Submission thread is now published as
Submission summary
| Authors (as registered SciPost users): | Davide Dreon · Lorenzo Fioroni |
| Submission information | |
|---|---|
| Preprint Link: | https://arxiv.org/abs/2404.14401v3 (pdf) |
| Code repository: | https://github.com/qo-eth/TorchGPE |
| Date accepted: | 2024-10-08 |
| Date submitted: | 2024-09-04 11:45 |
| Submitted by: | Dreon, Davide |
| Submitted to: | SciPost Physics Codebases |
| Ontological classification | |
|---|---|
| Academic field: | Physics |
| Specialties: |
|
| Approach: | Computational |
Abstract
TorchGPE is a general-purpose Python package developed for solving the Gross-Pitaevskii equation (GPE). This solver is designed to integrate wave functions across a spectrum of linear and non-linear potentials. A distinctive aspect of TorchGPE is its modular approach, which allows the incorporation of arbitrary self-consistent and time-dependent potentials, e.g., those relevant in many-body cavity QED models. The package employs a symmetric split-step Fourier propagation method, effective in both real and imaginary time. In our work, we demonstrate a significant improvement in computational efficiency by leveraging GPU computing capabilities. With the integration of the latter technology, TorchGPE achieves a substantial speed-up with respect to conventional CPU-based methods, greatly expanding the scope and potential of research in this field.
Author comments upon resubmission
Thank you for your positive assessment of the paper and for providing us with very constructive and useful reports.
We are pleased to read that both Referees liked the pedagogical style of the paper and the variety of problems our code could help solve with its GPU speedup. The Referees also agreed on the usefulness of additional benchmarks. We have followed their suggestions and added a use case, while more examples and details of how to use the package have been kept in the software documentation to make the paper easier to read.
We have already included a point-by-point response to each of the referees' reports. In the "list of changes" below is a summary of the main changes made to the paper.
We believe that we have addressed all their recommendations and we hope that this version is suitable for publication in SciPost Physics Codebases.
With kind regards,
Davide Dreon on behalf of all authors of the manuscript
List of changes
- In Section 4.1, highlighted the key characteristics of a GPU that most significantly affect the performance of the code
- Added a comparison with another library in Section 4.1, to highlight the performance improvements. Due to its popularity, we have chosen to compare our library to GPELab
- Added the additional benchmark of a harmonically trapped BEC (section 5.1)
- Added comments on benchmarks convergence
- Reference to our previous experimental work where a preliminary version of TorchGPE had been used
- Minor corrections and specifications (particle numbers used, grid sizes, kappa definition) as suggested by the Referees
Published as SciPost Phys. Codebases 38 (2024) , SciPost Phys. Codebases 38-r1.0 (2024)
Reports on this Submission
Report
The revised manuscript has satisfactorily addressed the points raised in the report. I thank the authors for their reply and recommend the revised manuscript for acceptance.
Recommendation
Publish (easily meets expectations and criteria for this Journal; among top 50%)
Strengths
See previous report
Weaknesses
See previous report
Report
The authors have addressed both my and the other referees comments on their manscript fully. It is is now suitable for publication.
Requested changes
None
Recommendation
Publish (meets expectations and criteria for this Journal)