EXaCTz: Guaranteed Extremum Graph and Contour Tree Preservation for Distributed- and GPU-Parallel Lossy Compression
Yuxiao Li, Mingze Xia, Xin Liang, Bei Wang, and Hanqi Guo

TL;DR
EXaCTz is a high-performance, topology-preserving lossy compression algorithm for scalar fields that guarantees convergence and significantly outperforms existing methods in speed and scalability.
Contribution
It introduces a novel, bounded-iteration algorithm that enforces topological consistency during lossy compression, with theoretical convergence guarantees and scalable GPU and distributed implementations.
Findings
Achieves up to 4.52 GB/s throughput on a single GPU.
Outperforms state-of-the-art methods by up to 213x in speed.
Scales to 128 GPUs, processing 512 GB datasets in under 48 seconds.
Abstract
This paper introduces EXaCTz, a parallel algorithm that concurrently preserves extremum graphs and contour trees in lossy-compressed scalar field data. While error-bounded lossy compression is essential for large-scale scientific simulations and workflows, existing topology-preserving methods suffer from (1) a significant throughput disparity, where topology correction speeds are on the order of MB/s, lagging orders of magnitude behind compression speeds on the order of GB/s, (2) limited support for diverse topological descriptors, and (3) a lack of theoretical convergence bounds. To address these challenges, EXaCTz introduces a high-performance, bounded-iteration algorithm that enforces topological consistency by deriving targeted edits for decompressed data. Unlike prior methods that rely on explicit topology reconstruction, EXaCTz enforces consistent min/max neighbors of all…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
