GPU-based Image Compression for Efficient Compositing in Distributed Rendering Applications

Riley Lipinksi; Kenneth Moreland; Michael E. Papka; Thomas Marrinan

doi:10.1109/LDAV53230.2021.00012

GPU-based Image Compression for Efficient Compositing in Distributed Rendering Applications

Riley Lipinksi, Kenneth Moreland, Michael E. Papka, Thomas Marrinan

Visualization

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Visualizations of large-scale data sets are often created on graphics clusters that distribute the rendering task amongst many processes. When using real-time GPU-based graphics algorithms, the most time-consuming aspect of distributed rendering is typically the com-positing phase - combining all partial images from each rendering process into the final visualization. Compo siting requires image data to be copied off the GPU and sent over a network to other processes. While compression has been utilized in existing distributed rendering compositors to reduce the data being sent over the network, this compression tends to occur after the raw images are transferred from the GPU to main memory. In this paper, we present work that leverages OpenGL / CUDA interoperability to compress raw images on the GPU prior to transferring the data to main memory. This approach can significantly reduce the device-to-host data transfer time, thus enabling more efficient compositing of images generated by distributed rendering applications.

Original language	English
Title of host publication	Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	43-52
Number of pages	10
ISBN (Electronic)	9781665432832
DOIs	https://doi.org/10.1109/LDAV53230.2021.00012
State	Published - 2021
Event	11th IEEE Symposium on Large Data Analysis and Visualization, LDAV 2021 - Virtual, Online, United States Duration: Oct 25 2021 → …

Publication series

Name	Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021

Conference

Conference	11th IEEE Symposium on Large Data Analysis and Visualization, LDAV 2021
Country/Territory	United States
City	Virtual, Online
Period	10/25/21 → …

Funding

We would like to thank Mykhailo Ohorodnichuk and TurboSquid for creating and providing the “3 D Nuclear station Low-poly” model, Narayanan Kasthuri (Argonne National Laboratory and the University of Chicago) for providing the data for the reconstructed neuron models, and OpenStreetMap for collecting GPS coordinate data and making it publicly available. This research was supported in part by the Argonne Leadership Computing Facility and the Oak Ridge Leadership Computing Facility, which are U.S. Department of Energy Office of Science User Facilities operated under contracts DE-AC02-06CH11357 and DE-AC05-00OR22725 respectively.

Keywords

Computer graphics-Image compression
Computing methodologies
Massively parallel algorithms
Parallel algorithms
Parallel computing methodologies

Access to Document

10.1109/LDAV53230.2021.00012

Cite this

Lipinksi, R., Moreland, K., Papka, M. E., & Marrinan, T. (2021). GPU-based Image Compression for Efficient Compositing in Distributed Rendering Applications. In Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021 (pp. 43-52). (Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/LDAV53230.2021.00012

Lipinksi, Riley ; Moreland, Kenneth ; Papka, Michael E. et al. / GPU-based Image Compression for Efficient Compositing in Distributed Rendering Applications. Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021. Institute of Electrical and Electronics Engineers Inc., 2021. pp. 43-52 (Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021).

@inproceedings{678b8b40362d4d8aa70a601ab95f9b41,

title = "GPU-based Image Compression for Efficient Compositing in Distributed Rendering Applications",

abstract = "Visualizations of large-scale data sets are often created on graphics clusters that distribute the rendering task amongst many processes. When using real-time GPU-based graphics algorithms, the most time-consuming aspect of distributed rendering is typically the com-positing phase - combining all partial images from each rendering process into the final visualization. Compo siting requires image data to be copied off the GPU and sent over a network to other processes. While compression has been utilized in existing distributed rendering compositors to reduce the data being sent over the network, this compression tends to occur after the raw images are transferred from the GPU to main memory. In this paper, we present work that leverages OpenGL / CUDA interoperability to compress raw images on the GPU prior to transferring the data to main memory. This approach can significantly reduce the device-to-host data transfer time, thus enabling more efficient compositing of images generated by distributed rendering applications.",

keywords = "Computer graphics-Image compression, Computing methodologies, Massively parallel algorithms, Parallel algorithms, Parallel computing methodologies",

author = "Riley Lipinksi and Kenneth Moreland and Papka, {Michael E.} and Thomas Marrinan",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 11th IEEE Symposium on Large Data Analysis and Visualization, LDAV 2021 ; Conference date: 25-10-2021",

year = "2021",

doi = "10.1109/LDAV53230.2021.00012",

language = "English",

series = "Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "43--52",

booktitle = "Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021",

}

Lipinksi, R, Moreland, K, Papka, ME & Marrinan, T 2021, GPU-based Image Compression for Efficient Compositing in Distributed Rendering Applications. in Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021. Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021, Institute of Electrical and Electronics Engineers Inc., pp. 43-52, 11th IEEE Symposium on Large Data Analysis and Visualization, LDAV 2021, Virtual, Online, United States, 10/25/21. https://doi.org/10.1109/LDAV53230.2021.00012

GPU-based Image Compression for Efficient Compositing in Distributed Rendering Applications. / Lipinksi, Riley; Moreland, Kenneth; Papka, Michael E. et al.
Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021. Institute of Electrical and Electronics Engineers Inc., 2021. p. 43-52 (Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - GPU-based Image Compression for Efficient Compositing in Distributed Rendering Applications

AU - Lipinksi, Riley

AU - Moreland, Kenneth

AU - Papka, Michael E.

AU - Marrinan, Thomas

PY - 2021

Y1 - 2021

N2 - Visualizations of large-scale data sets are often created on graphics clusters that distribute the rendering task amongst many processes. When using real-time GPU-based graphics algorithms, the most time-consuming aspect of distributed rendering is typically the com-positing phase - combining all partial images from each rendering process into the final visualization. Compo siting requires image data to be copied off the GPU and sent over a network to other processes. While compression has been utilized in existing distributed rendering compositors to reduce the data being sent over the network, this compression tends to occur after the raw images are transferred from the GPU to main memory. In this paper, we present work that leverages OpenGL / CUDA interoperability to compress raw images on the GPU prior to transferring the data to main memory. This approach can significantly reduce the device-to-host data transfer time, thus enabling more efficient compositing of images generated by distributed rendering applications.

AB - Visualizations of large-scale data sets are often created on graphics clusters that distribute the rendering task amongst many processes. When using real-time GPU-based graphics algorithms, the most time-consuming aspect of distributed rendering is typically the com-positing phase - combining all partial images from each rendering process into the final visualization. Compo siting requires image data to be copied off the GPU and sent over a network to other processes. While compression has been utilized in existing distributed rendering compositors to reduce the data being sent over the network, this compression tends to occur after the raw images are transferred from the GPU to main memory. In this paper, we present work that leverages OpenGL / CUDA interoperability to compress raw images on the GPU prior to transferring the data to main memory. This approach can significantly reduce the device-to-host data transfer time, thus enabling more efficient compositing of images generated by distributed rendering applications.

KW - Computer graphics-Image compression

KW - Computing methodologies

KW - Massively parallel algorithms

KW - Parallel algorithms

KW - Parallel computing methodologies

UR - http://www.scopus.com/inward/record.url?scp=85123959837&partnerID=8YFLogxK

U2 - 10.1109/LDAV53230.2021.00012

DO - 10.1109/LDAV53230.2021.00012

M3 - Conference contribution

AN - SCOPUS:85123959837

T3 - Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021

SP - 43

EP - 52

BT - Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 11th IEEE Symposium on Large Data Analysis and Visualization, LDAV 2021

Y2 - 25 October 2021

ER -

Lipinksi R, Moreland K, Papka ME, Marrinan T. GPU-based Image Compression for Efficient Compositing in Distributed Rendering Applications. In Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021. Institute of Electrical and Electronics Engineers Inc. 2021. p. 43-52. (Proceedings - 2021 IEEE 11th Symposium on Large Data Analysis and Visualization, LDAV 2021). doi: 10.1109/LDAV53230.2021.00012

GPU-based Image Compression for Efficient Compositing in Distributed Rendering Applications

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this