A significant challenge on an exascale computer is the speed at which we compute results exceeds by many orders of magnitude the speed at which we save these results. Therefore the Exascale Computing Project (ECP) ALPINE project focuses on providing exascale-ready visualization solutions including in situ processing. In situ visualization and analysis runs as the simulation is run, on simulations results are they are generated avoiding the need to save entire simulations to storage for later analysis. The ALPINE project made post hoc visualization tools, ParaView and VisIt, exascale ready and developed in situ algorithms and infrastructures. The suite of ALPINE algorithms developed under ECP includes novel approaches to enable automated data analysis and visualization to focus on the most important aspects of the simulation. Many of the algorithms also provide data reduction benefits to meet the I/O challenges at exascale. ALPINE developed a new lightweight in situ infrastructure, Ascent.

Contour trees offer an abstract representation of the level set topology in scalar fields and are widely used in topological data analysis and visualization. However, applying contour trees to large-scale scientific datasets remains challenging due to scalability limitations. Recent developments in distributed hierarchical contour trees have addressed these challenges by enabling scalable computation across distributed systems. Building on these structures, advanced analytical tasks—such as volumetric branch decomposition and contour extraction—have been introduced to facilitate large-scale scientific analysis. Despite these advancements, such analytical tasks substantially increase memory usage, which hampers scalability. In this paper, we propose a pre-simplification strategy to significantly reduce the memory overhead associated with analytical tasks on distributed hierarchical contour trees. We demonstrate enhanced scalability through strong scaling experiments, constructing the largest known contour tree—comprising over half a trillion nodes with complex topology—in under 15 minutes on a dataset containing 550 billion elements.

We present the first combinatorial algorithm for efficiently computing the Reeb space in all dimensions. The Reeb space is a higher-dimensional generalization of the Reeb graph, which is standard practice in the analysis of scalar fields, along with other computational topology tools such as persistent homology and the Morse-Smale complex. One significant limitation of topological tools for scalar fields is that data often involves multiple variables, where joint analysis is more insightful. Generalizing topological data structures to multivariate data has proven challenging and the Reeb space is one of the few available options. However, none of the existing algorithms can efficiently compute the Reeb space in arbitrary dimensions and there are no available implementations which are robust with respect to numerical errors. We propose a new algorithm for computing the Reeb space of a generic piecewise linear map over a simplicial mesh of any dimension called arrange and traverse. We implement a robust specialization of our algorithm for tetrahedral meshes and evaluate it on real-life data.

Many scientific and engineering problems are modelled by simulating scalar fields defined either on space-filling meshes (Eulerian) or as particles (Lagrangian). For analysis and visualization, topological primitives such as contour trees can be used, but these often need simplification to filter out small-scale features. For parcel-based convective cloud simulations, simplification of the contour tree requires a volumetric measure rather than persistence. Unlike for cubic meshes, volume cannot be approximated by counting regular vertices. Typically, this is addressed by resampling irregular data onto a uniform grid. Unfortunately, the spatial proximity of parcels requires a high sampling frequency, resulting in a massive increase in data size for processing. We therefore extend volumebased contour tree simplification to parcel-in-cell simulations with a graph adaptor in Viskores ( VTK-m), using Delaunay tetrahedralization of the parcel centroids as input. Instead of relying on a volume approximation by counting regular vertices - as was done for cubic meshes - we adapt the 2D area splines reported by Bajaj et al. [1], and Zhou et al. [37]. We implement this in Viskores (formerly called VTK-m) as prefix-sum style hypersweeps for parallel efficiency and show how it can be generalized to compute any integrable property. Finally, our results reveal that contour trees computed directly on the parcels are orders of magnitude faster than computing them on a resampled grid, while also arguably offering better quality segmentation, avoiding interpolation artifacts.

Feature level sets (FLS) have shown significant potential in the analysis of multi-field data by using traits defined in attribute space to specify features in the domain. In this work, we address key challenges in the practical use of FLS: trait design and feature selection for rendering. To simplify trait design, we propose a Cartesian decomposition of traits into simpler components, making the process more intuitive and computationally efficient. Additionally, we utilize dictionary learning results to automatically suggest point traits. To enhance feature selection, we introduce trait-induced merge trees (TIMTs), a generalization of merge trees for feature level sets, aimed at topologically analyzing tensor fields or general multi-variate data. The leaves in the TIMT represent areas in the input data that are closest to the defined trait, thereby most closely resembling the defined feature. This merge tree provides a hierarchy of features, enabling the querying of the most relevant and persistent features. Our method includes various query techniques for the tree, allowing the highlighting of different aspects. We demonstrate the cross-application capabilities of this approach through five case studies from different domains.

Contour trees describe the topology of level sets in scalar fields and are widely used in topological data analysis and visualization. A main challenge of utilizing contour trees for large-scale scientific data is their computation at scale using high-performance computing. To address this challenge, recent work has introduced distributed hierarchical contour trees for distributed computation and storage of contour trees. However, effective use of these distributed structures in analysis and visualization requires subsequent computation of geometric properties and branch decomposition to support contour extraction and exploration. In this work, we introduce distributed algorithms for augmentation, hypersweeps, and branch decomposition that enable parallel computation of geometric properties, and support the use of distributed contour trees as query structures for scientific exploration. We evaluate the parallel performance of these algorithms and apply them to identify and extract important contours for scientific visualization.

Africa is poised for a revolution in the quality and relevance of weather predictions, with potential for great benefits in terms of human and economic security. This revolution will be driven by recent international progress in nowcasting, numerical weather prediction, theoretical tropical dynamics, and forecast communication, but will depend on suitable scientific investment being made. The commercial sector has recognized this opportunity and new forecast products are being made available to African stakeholders. At this time, it is vital that robust scientific methods are used to develop and evaluate the new generation of forecasts. The Global Challenges Research Fund (GCRF) African Science for Weather Information and Forecasting Techniques (SWIFT) project represents an international effort to advance scientific solutions across the fields of nowcasting, synoptic and short-range severe weather prediction, subseasonal-to-seasonal (S2S) prediction, user engagement, and forecast evaluation. This paper describes the opportunities facing African meteorology and the ways in which SWIFT is meeting those opportunities and identifying priority next steps. Delivery and maintenance of weather forecasting systems exploiting these new solutions requires a trained body of scientists with skills in research and training, modeling and operational prediction, and communications and leadership. By supporting partnerships between academia and operational agencies in four African partner countries, the SWIFT project is helping to build capacity and capability in African forecasting science. A highlight of SWIFT is the coordination of three weather forecasting “Testbeds”—the first of their kind in Africa—which have been used to bring new evaluation tools, research insights, user perspectives, and communications pathways into a semioperational forecasting environment.

Contour trees are used for topological data analysis in scientific visualization. While originally computed with serial algorithms, recent work has introduced a vector-parallel algorithm. However, this algorithm is relatively slow for fully augmented contour trees which are needed for many practical data analysis tasks. We therefore introduce a representation called the hyperstructure that enables efficient searches through the contour tree and use it to construct a fully augmented contour tree in data parallel, with performance on average 6 times faster than the state-of-the-art parallel algorithm in the TTK topological toolkit.

Contour trees are a significant tool for data analysis as they capture both local and global variation. However, their utility has been limited by scalability, in particular for distributed computation and storage. We report a distributed data structure for storing the contour tree of a data set distributed on a cluster, based on a fan-in hierarchy, and an algorithm for computing it based on the boundary tree that represents only the superarcs of a contour tree that involve contours that cross boundaries between blocks. This allows us to limit the communication cost for contour tree computation to the complexity of the block boundaries rather than of the entire data set.

The contour tree is one of the principal tools in scientific visualisation. It captures the connectivity of level sets in scalar fields. In order to apply the contour tree to exascale data we need efficient shared memory and distributed algorithms. Recent work has revealed a parallel performance bottleneck caused by substructures of contour trees called W-structures. We report two novel algorithms that detect and extract the W-structures. We also use the W-structures to show that extended persistence is not equivalent to branch decomposition and leaf-pruning.

As data sets grow to exascale, automated data analysis and visualization are increasingly important, to intermediate human understanding and to reduce demands on disk storage via in situ analysis. Trends in architecture of high performance computing systems necessitate analysis algorithms to make effective use of combinations of massively multicore and distributed systems. One of the principal analytic tools is the contour tree, which analyses relationships between contours to identify features of more than local importance. Unfortunately, the predominant algorithms for computing the contour tree are explicitly serial, and founded on serial metaphors, which has limited the scalability of this form of analysis. While there is some work on distributed contour tree computation, and separately on hybrid GPU-CPU computation, there is no efficient algorithm with strong formal guarantees on performance allied with fast practical performance. We report the first shared SMP algorithm for fully parallel contour tree computation, with formal guarantees of O(lg V lgt) parallel steps and O(V lg V) work for data with V samples and t contour tree supernodes, and implementations with more than 30× parallel speed up on both CPU using TBB and GPU using Thrust and up 70× speed up compared to the serial sweep and merge algorithm.

As Exascale computing proliferates, we see an accelerating shift towards clusters with thousands of nodes and thousands of cores per node, often on the back of commodity graphics processing units. This paper argues that this drives a once in a generation shift of computation, and that fundamentals of computer science therefore need to be re-examined. Exploiting the full power of Exascale computation will require attention to the fundamentals of programme design and specification, programming language design, systems and software engineering, analytic, performance and cost models, fundamental algorithmic design, and to increasing replacement of human bandwidth by computational analysis. As part of this, we will argue that Exascale computing will require a significant degree of co-design and close attention to the economics underlying the challenges ahead.

The contour tree is a tool for understanding the topological structure of a scalar field. Recent work has built efficient contour tree algorithms for shared memory parallel computation, driven by the need to analyze large data sets in situ while the simulation is running. Unfortunately, methods for using the contour tree for practical data analysis are still primarily serial, including single isocontour extraction, branch decomposition and simplification. We report data parallel methods for these tasks using a data structure called the hyperstructure and a general purpose approach called a hypersweep. We implement and integrate these methods with a Cinema database that stores features as depth images and with a web server that reconstructs the features for direct visualization.

The fiber surface generalizes the popular isosurface to multi-fields, so that pre-images can be visualized as surfaces. As with the isosurface, however, the fiber surface suffers from visual occlusion. We propose to avoid such occlusion by restricting the components to only the relevant ones with a new component-wise flexing algorithm. The approach, flexible fiber surface, generalizes the manipulation idea found in the flexible isosurface for the fiber surface. The flexible isosurface in the original form, however, relies on the contour tree. For the fiber surface, this corresponds to the Reeb space, which is challenging for both the computation and user interaction. We thus take a Reeb-free approach, in which one does not compute the Reeb space. Under this constraint, we generalize a few selected interactions in the flexible isosurface and discuss the implication of the restriction.