IEEE VIS 2024 Content: TTK is Getting MPI-Ready

TTK is Getting MPI-Ready

E. Le Guillou -

M. Will -

P. Guillou -

J. Lukasczyk -

P. Fortin -

C. Garth -

J. Tierny -

Screen-reader Accessible PDF
Download preprint PDF
DOI: 10.1109/TVCG.2024.3390219

Room: Bayshore I

2024-10-17T16:12:00ZGMT-0600Change your timezone on the schedule page
2024-10-17T16:12:00Z
Exemplar figure, described by caption below
Output of an integrated pipeline that produces a real-life use case combining all of the algorithms parallelized in our paper. The pipeline is executed on the Turbulent Channel Flow dataset (120 billion vertices), a three-dimensional regular grid with two scalar fields, the pressure of the fluid and its gradient magnitude. The spheres correspond to the pressure critical points and the tubes are the integral lines starting at saddle points. Figure (a) shows all of the produced geometry, while (b) and (c) show parts of the output zoomed in.
Fast forward
Full Video
Keywords

Topological data analysis, high-performance computing, distributed-memory algorithms.

Abstract

This system paper documents the technical foundations for the extension of the Topology ToolKit (TTK) to distributed-memory parallelism with the Message Passing Interface (MPI). While several recent papers introduced topology-based approaches for distributed-memory environments, these were reporting experiments obtained with tailored, mono-algorithm implementations. In contrast, we describe in this paper a versatile approach (supporting both triangulated domains and regular grids) for the support of topological analysis pipelines, i.e. a sequence of topological algorithms interacting together, possibly on distinct numbers of processes. While developing this extension, we faced several algorithmic and software engineering challenges, which we document in this paper. We describe an MPI extension of TTK’s data structure for triangulation representation and traversal, a central component to the global performance and generality of TTK’s topological implementations. We also introduce an intermediate interface between TTK and MPI, both at the global pipeline level, and at the fine-grain algorithmic level. We provide a taxonomy for the distributed-memory topological algorithms supported by TTK, depending on their communication needs and provide examples of hybrid MPI+thread parallelizations. Detailed performance analyses show that parallel efficiencies range from 20% to 80% (depending on the algorithms), and that the MPI-specific preconditioning introduced by our framework induces a negligible computation time overhead. We illustrate the new distributed-memory capabilities of TTK with an example of advanced analysis pipeline, combining multiple algorithms, run on the largest publicly available dataset we have found (120 billion vertices) on a standard cluster with 64 nodes (for a total of 1536 cores). Finally, we provide a roadmap for the completion of TTK’s MPI extension, along with generic recommendations for each algorithm communication category.