Skip to Main content Skip to Navigation
Conference papers

A GPU approach to distance geometry in 1D: an implementation in C/CUDA

Abstract : We present a GPU implementation in C and CUDA of a matrix-by-vector procedure that is particularly tailored to a special class of distance geometry problems in dimension 1, which we name "paradoxical DGP instances". This matrix-byvector reformulation was proposed in previous studies on an optical processor specialized for this kind of computations. Our computational experiments show that a consistent speed-up is observed when comparing our GPU implementation against a standard algorithm for distance geometry, called the Branchand-Prune algorithm. These results confirm that a suitable implementation of the matrix-by-vector procedure in the context of optic computing is very promising. We also remark, however, that the total number of detected solutions grows with the instance size in our implementations, which appears to be an important limitation to the effective implementation of the optical processor.
Complete list of metadata

https://hal.inria.fr/hal-03746879
Contributor : Antonio Mucherino Connect in order to contact the contributor
Submitted on : Monday, August 8, 2022 - 3:05:25 PM
Last modification on : Thursday, August 11, 2022 - 3:03:31 AM

File

bp1_on_GPU_HAL_Extended.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03746879, version 2

Citation

Simon Hengeveld, Antonio Mucherino. A GPU approach to distance geometry in 1D: an implementation in C/CUDA. 17th Conference on Computer Science and Intelligence Systems, Sep 2022, Sofia, Bulgaria. ⟨hal-03746879v2⟩

Share

Metrics

Record views

13

Files downloads

10