1 of 1

Asynchronous and Load-Balanced Union-Find for Distributed and Parallel Scientific Data Visualization and Analysis

Significance and Impact

This paper won the best paper award at IEEE Pacific Visualization Symposium 2021. This new technique achieves over 20x speedup over the previous state-of-the-art, with benchmarks on high-speed imaging experimental data and fusion plasma simulations using 1,024 processors.

Fig. 1: Strong scaling of distributed union-find for tracking and extracting features in two application datasets.

J. Xu, H. Guo, H.-W. Shen, M. Raj, X. Wang, X. Xu, Z. Wang, T. Peterka, IEEE Transactions on Visualization and Computer Graphics 27 (6): 2808-2820 (2021).

Scientific Achievement

This study presents a novel distributed union-find algorithm that features asynchronous parallelism and k-d tree based load balancing for scalable visualization and analysis of scientific data.

Research Details

• Proved that global synchronizations in existing distributed union-find can be eliminated without changing final results, allowing overlapped communications and computations for scalable processing.
• Leveraged a k-d tree decomposition to redistribute inputs, in order to improve workload balancing.