Parallel DTFE surface density field reconstruction

Esteban Rangel, Nan Li, Salman Habib, Tom Peterka, Ankit Agrawal, Wei-Keng Liao, Alok Nidhi Choudhary

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

We improve the interpolation accuracy and efficiency of the Delaunay tessellation field estimator (DTFE) for surface density field reconstruction by proposing an algorithm that takes advantage of the adaptive triangular mesh for lineof- sight integration. The costly computation of an intermediate 3D grid is completely avoided by our method and only optimally chosen interpolation points are computed, thus, the overall computational cost is significantly reduced. The algorithm is implemented as a parallel shared-memory kernel for large-scale grid rendered field reconstructions in our distributed-memory framework designed for N-body gravitational lensing simulations in large volumes. We also introduce a load balancing scheme to optimize the efficiency of processing a large number of field reconstructions. Our results show our kernel outperforms existing software packages for volume weighted density field reconstruction, achieving ∼10× speedup, and our load balancing algorithm gains an additional ∼3.6× speedup at scales with ∼16k processes.

Original languageEnglish (US)
Title of host publicationProceedings - 2016 IEEE International Conference on Cluster Computing, CLUSTER 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages30-39
Number of pages10
ISBN (Electronic)9781509036530
DOIs
StatePublished - Dec 6 2016
Event2016 IEEE International Conference on Cluster Computing, CLUSTER 2016 - Taipei, Taiwan, Province of China
Duration: Sep 13 2016Sep 15 2016

Other

Other2016 IEEE International Conference on Cluster Computing, CLUSTER 2016
CountryTaiwan, Province of China
CityTaipei
Period9/13/169/15/16

Keywords

  • Delaunay tessellation field estimator
  • Parallel surface density

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture
  • Signal Processing

Fingerprint Dive into the research topics of 'Parallel DTFE surface density field reconstruction'. Together they form a unique fingerprint.

  • Cite this

    Rangel, E., Li, N., Habib, S., Peterka, T., Agrawal, A., Liao, W-K., & Choudhary, A. N. (2016). Parallel DTFE surface density field reconstruction. In Proceedings - 2016 IEEE International Conference on Cluster Computing, CLUSTER 2016 (pp. 30-39). [7776476] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CLUSTER.2016.40