Saved in:
Bibliographic Details
Main Authors: Wu, Qi, Insley, Joseph A., Mateevitsi, Victor A., Rizzi, Silvio, Papka, Michael E., Ma, Kwan-Liu
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2304.10516
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913438488330240
author Wu, Qi
Insley, Joseph A.
Mateevitsi, Victor A.
Rizzi, Silvio
Papka, Michael E.
Ma, Kwan-Liu
author_facet Wu, Qi
Insley, Joseph A.
Mateevitsi, Victor A.
Rizzi, Silvio
Papka, Michael E.
Ma, Kwan-Liu
contents Implicit neural representations (INRs) have emerged as a powerful tool for compressing large-scale volume data. This opens up new possibilities for in situ visualization. However, the efficient application of INRs to distributed data remains an underexplored area. In this work, we develop a distributed volumetric neural representation and optimize it for in situ visualization. Our technique eliminates data exchanges between processes, achieving state-of-the-art compression speed, quality and ratios. Our technique also enables the implementation of an efficient strategy for caching large-scale simulation data in high temporal frequencies, further facilitating the use of reactive in situ visualization in a wider range of scientific problems. We integrate this system with the Ascent infrastructure and evaluate its performance and usability using real-world simulations.
format Preprint
id arxiv_https___arxiv_org_abs_2304_10516
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Distributed Neural Representation for Reactive in situ Visualization
Wu, Qi
Insley, Joseph A.
Mateevitsi, Victor A.
Rizzi, Silvio
Papka, Michael E.
Ma, Kwan-Liu
Distributed, Parallel, and Cluster Computing
Artificial Intelligence
Implicit neural representations (INRs) have emerged as a powerful tool for compressing large-scale volume data. This opens up new possibilities for in situ visualization. However, the efficient application of INRs to distributed data remains an underexplored area. In this work, we develop a distributed volumetric neural representation and optimize it for in situ visualization. Our technique eliminates data exchanges between processes, achieving state-of-the-art compression speed, quality and ratios. Our technique also enables the implementation of an efficient strategy for caching large-scale simulation data in high temporal frequencies, further facilitating the use of reactive in situ visualization in a wider range of scientific problems. We integrate this system with the Ascent infrastructure and evaluate its performance and usability using real-world simulations.
title Distributed Neural Representation for Reactive in situ Visualization
topic Distributed, Parallel, and Cluster Computing
Artificial Intelligence
url https://arxiv.org/abs/2304.10516