Epidemic and pandemic preparedness with rapid outbreak response rely on timely, trustworthy evidence. Mathematical models are crucial for supporting timely and reliable evidence generation for public health decision-making with models spanning approaches from compartmental and metapopulation models to detailed agent-based simulations. Yet, the accompanying software ecosystem remains fragmented across model types, spatial resolutions, and computational targets, making models harder to compare, extend, and deploy at scale. Here we present MEmilio, a modular, high-performance framework for epidemic simulation that harmonizes the specification and execution of diverse dynamic epidemiological models within a unified and harmonized architecture. MEmilio couples an efficient C++ simulation core with coherent model descriptions and a user-friendly Python interface, enabling workflows that run on laptops as well as high-performance computing systems. Standardized representations of space, demography, and mobility support straightforward adaptations in resolution and population size, facilitating systematic inter-model comparisons and ensemble studies. The framework integrates readily with established tools for uncertainty quantification and parameter inference, supporting a broad range of applications from scenario exploration to calibration. Finally, strict software-engineering practices, including extensive unit and continuous integration testing, promote robustness and minimize the risk of errors as the framework evolves. By unifying implementations across modeling paradigms, MEmilio aims to lower barriers to reuse and generalize models, enable principled comparisons of implicit assumptions, and accelerate the development of novel approaches that strengthen modeling-based outbreak preparedness.

Adequate Earth system simulations require interactions between the atmosphere, the ocean, and the ice sheets. To this end, numerical solvers that compute the evolution of the different Earth system components are coupled. There are frameworks and libraries for coupling that handle the complex tasks of coordinating solver execution, communicating between processes, and mapping between different meshes. This allows solvers to be developed independently without compromises on numerical methods or technology. Code reuse is improved, both over large, monolithic software systems that reimplement each coupled model as well as over ad-hoc coupling scripts. In this work, we use the preCICE coupling library to couple the Ice-sheet and Sea-level System Model (ISSM) with the subglacial hydrology model CUAS-MPI. An adapter for each model is required to pass meshes and coupled variables between the model and preCICE. We focus mainly on the technical aspects (design, development, and use of the adapters, choice of coupling library, and large-scale performance analysis), using a synthetic setup to verify functionality and correctness. The adapters we developed are generic and reusable for use cases other than ice-hydrology coupling. Computational performance for the coupled system is measured on a high-performance computing cluster. We find that coupling with preCICE has low computational overhead and does not negatively impact scaling. A comparison between unidirectional and bidirectional coupling for the synthetic ice sheet shows that the coupling captures the anticipated feedback mechanisms between the two systems. The coupled simulations are numerically stable, despite the nonlinearities in the physical system. The generic coupling library preCICE is well suited for our use case and has advantages as well as disadvantages over Earth System Model-specific libraries. The new framework and code enable studies of the subglacial hydrological systems of ice sheets, as well as coupling ISSM or CUAS-MPI with other codes, such as in global Earth System Models or process models.

Infectious diseases remain a major threat to human societies. During the recent COVID-19 pandemic, mathematical modeling and extensive computer simulations proved highly effective in supporting public health experts and decision makers. Despite these advances, the full potential of modern modeling approaches and digital technologies has not yet been realized. Many critical tasks – including expert consultations, model execution, scenario analyses, report preparation, and result communication – still relied heavily on manual, human-driven processes with each manual interaction introducing avoidable delays and limiting responsiveness during rapidly evolving outbreaks. Pandemic preparedness should opt for automated workflows and seamlessly integrated software modules that can improve pandemic mitigation capabilities by substantially reducing response times. For this step, we require robust and flexible computational infrastructure capable of supporting heterogeneous hardware and continuously evolving infectious-disease models. In addition, data sources need to be dynamically integrated. Managing such demands needs infrastructure that supports automated high-performance computing (HPC) workflows. Beyond computational performance, software infrastructure must ensure secure user and data management to comply with data-protection regulations and provide clear, transparent presentation of results to both decision makers and the public. Meeting the aforementioned challenges requires tight integration of state-of-the-art scientific software with modern, scalable infrastructure that can leverage supercomputing resources when necessary. For rapid deployment in future epidemic or pandemic scenarios, adherence to the FAIR principles for research software is critical to ensure reusability and sustainability.

Large-scale Earth observation (EO) datasets are crucial for identifying environmental changes, particularly in coastal regions vulnerable to erosion. However, analyzing these massive datasets requires computational techniques beyond the capabilities of conventional workstations or single nodes of an HPC system. In this study, we demonstrate how Heat, a highly parallel, open-source, Python-based library designed for scalable machine learning and data processing, can address this issue-quasi as an off-the-shelf solution for up-scaling-when detecting anomalies in satellite imagery of the German North Sea coast. Our implementation of an unsupervised outlier-based anomaly detection algorithm uses Heat's distributed arrays and vectorized map (vmap) primitives. This algorithm processes tens of millions of shoreline locations by sharing memory and computation across multi-node CPU/GPU clusters, with minimal code changes compared to NumPy/PyTorch. Weak-scaling experiments on both CPUs and GPUs demonstrate the scalability of our approach with increasing amounts of data. We thereby provide, to the best of our knowledge, the first density-based anomaly detection on large-scale EO datasets in a multi-node setting, with proven portability to different hardware architectures (x86-and ARM-based CPUs) and vendors (Nvidia and AMD GPUs). The resulting anomaly maps align with known severe weather episodes, daily anomaly counts correlate with wind metrics from coastal stations, and hotspot maps identify regions of high activity. These findings support the geophysical plausibility of the detections. Our approach is reproducible due to its deterministic algorithms, extensible to additional EO modalities (e.g. coherence), and broadly applicable to nationwide monitoring. Although the 20 m data resolution limits detection sensitivity, the method itself is resolution-agnostic.

Large-scale earth system simulations produce huge amounts of data. Due to limited I/O bandwidth and available storage space this data often needs to be reduced before writing to disk or storing permanently. Error-bounded lossy compression is an effective approach to tackle the trade-off between accuracy and storage space. We are exploring and discussing error-bounded lossy compression based on tree-based adaptive mesh refinement (AMR) techniques. According to flexible error-criteria the simulation data is coarsened until a given error bound is reached. This reduces the number of mesh elements and data points significantly. The error criterion may for example be an absolute or relative point-wise error. Since the compression method is closely linked to the mesh we can additionally incorporate geometry information - for example varying the error by geospatial region. We implement these techniques as the open source tool cmc, which is based on the parallel AMR library t8code. The compression tool can be linked to and used by arbitrary simulation applications or executed as a post-processing step. As a first example, we couple our compressor with the MESSy and MPTRAC libraries. We show different results including the compression of ERA5 data. The compressed sample datasets show better results in terms of file size than conventional compressors such as SZ and ZFP. In addition, our method allows for a more fine-grained error control.

Recent studies have confirmed cardiovascular diseases remain responsible for highest death toll amongst non-communicable diseases. Accurate left ventricular (LV) volume estimation is critical for valid diagnosis and management of various cardiovascular conditions, but poses significant challenge due to inherent uncertainties associated with segmentation algorithms in magnetic resonance imaging (MRI). Recent machine learning advancements, particularly U-Net-like convolutional networks, have facilitated automated segmentation for medical images, but struggles under certain pathologies and/or different scanner vendors and imaging protocols. This study proposes a novel methodology for post-hoc uncertainty estimation in LV volume prediction using It\^{o} stochastic differential equations (SDEs) to model path-wise behavior for the prediction error. The model describes the area of the left ventricle along the heart's long axis. The method is agnostic to the underlying segmentation algorithm, facilitating its use with various existing and future segmentation technologies. The proposed approach provides a mechanism for quantifying uncertainty, enabling medical professionals to intervene for unreliable predictions. This is of utmost importance in critical applications such as medical diagnosis, where prediction accuracy and reliability can directly impact patient outcomes. The method is also robust to dataset changes, enabling application for medical centers with limited access to labeled data. Our findings highlight the proposed uncertainty estimation methodology's potential to enhance automated segmentation robustness and generalizability, paving the way for more reliable and accurate LV volume estimation in clinical settings as well as opening new avenues for uncertainty quantification in biomedical image segmentation, providing promising directions for future research.

Agent-based models have proven to be useful tools in supporting decision-making processes in different application domains. The advent of modern computers and supercomputers has enabled these bottom-up approaches to realistically model human mobility and contact behavior. The COVID-19 pandemic showcased the urgent need for detailed and informative models that can answer research questions on transmission dynamics. We present a sophisticated agent-based model to simulate the spread of respiratory diseases. The model is highly modularized and can be used on various scales, from a small collection of buildings up to cities or countries. Although not being the focus of this paper, the model has undergone performance engineering on a single core and provides an efficient intra- and inter-simulation parallelization for time-critical decision-making processes. In order to allow answering research questions on individual level resolution, nonpharmaceutical intervention strategies such as face masks or venue closures can be implemented for particular locations or agents. In particular, we allow for sophisticated testing and isolation strategies to study the effects of minimal-invasive infectious disease mitigation. With realistic human mobility patterns for the region of Brunswick, Germany, we study the effects of different interventions between March 1st and May 30, 2021 in the SARS-CoV-2 pandemic. Our analyses suggest that symptom-independent testing has limited impact on the mitigation of disease dynamics if the dark figure in symptomatic cases is high. Furthermore, we found that quarantine length is more important than quarantine efficiency but that, with sufficient symptomatic control, also short quarantines can have a substantial effect.

Tensor networks are a class of algorithms aimed at reducing the computational complexity of high-dimensional problems. They are used in an increasing number of applications, from quantum simulations to machine learning. Exploiting data parallelism in these algorithms is key to using modern hardware. However, there are several ways to map required tensor operations onto linear algebra routines (“building blocks”). Optimizing this mapping impacts the numerical behavior, so computational and numerical aspects must be considered hand-in-hand. In this paper we discuss the performance of solvers for low-rank linear systems in the tensor-train format (also known as matrix-product states). We consider three popular algorithms: TT-GMRES, MALS, and AMEn. We illustrate their computational complexity based on the example of discretizing a simple high-dimensional PDE in, for example, 5010 grid points. This shows that the projection to smaller sub-problems for MALS and AMEn reduces the number of floating-point operations by orders of magnitude. We suggest optimizations regarding orthogonalization steps, singular value decompositions, and tensor contractions. In addition, we propose a generic preconditioner based on a TT-rank-1 approximation of the linear operator. Overall, we obtain roughly a 5× speedup over the reference algorithm for the fastest method (AMEn) on a current multicore CPU.

Cardiac real-time magnetic resonance imaging (MRI) is an emerging technology that images the heart at up to 50 frames per second, offering insight into the respiratory effects on the heartbeat. However, this method significantly increases the number of images that must be segmented to derive critical health indicators. Although neural networks perform well on inner slices, predictions on outer slices are often unreliable. This work proposes sparse Bayesian learning (SBL) to predict the ventricular volume on outer slices with minimal manual labeling to address this challenge. The ventricular volume over time is assumed to be dominated by sparse frequencies corresponding to the heart and respiratory rates. Moreover, SBL identifies these sparse frequencies on well-segmented inner slices by optimizing hyperparameters via type -II likelihood, automatically pruning irrelevant components. The identified sparse frequencies guide the selection of outer slice images for labeling, minimizing posterior variance. This work provides performance guarantees for the greedy algorithm. Testing on patient data demonstrates that only a few labeled images are necessary for accurate volume prediction. The labeling procedure effectively avoids selecting inefficient images. Furthermore, the Bayesian approach provides uncertainty estimates, highlighting unreliable predictions (e.g., when choosing suboptimal labels).

The rapid advancements in quantum computing necessitate a scientific and rigorous approach to the construction of a corresponding software ecosystem, a topic underexplored and primed for systematic investigation. This chapter takes an important step in this direction: It presents scientific considerations essential for building a quantum software ecosystem that makes quantum computing available for scientific and industrial problem solving. Central to this discourse is the concept of hardware-software co-design, which fosters a bidirectional feedback loop from the application layer at the top of the software stack down to the hardware. This approach begins with compilers and low-level software that are specifically designed to align with the unique specifications and constraints of the quantum processor, proceeds with algorithms developed with a clear understanding of underlying hardware and computational model features, and extends to applications that effectively leverage the capabilities to achieve a quantum advantage. We analyze the ecosystem from two critical perspectives: the conceptual view, focusing on theoretical foundations, and the technical infrastructure, addressing practical implementations around real quantum devices necessary for a functional ecosystem. This approach ensures that the focus is towards promising applications with optimized algorithm-circuit synergy, while ensuring a user-friendly design, an effective data management and an overall orchestration. Our chapter thus offers a guide to the essential concepts and practical strategies necessary for developing a scientifically grounded quantum software ecosystem.

On the path to full understanding of the structure-function relationship or even design of RNA, structure prediction would offer an intriguing complement to experimental efforts. Any deep learning on RNA structure, however, is hampered by the sparsity of labeled training data. Utilizing the limited data available, we here focus on predicting spatial adjacencies ("contact maps”) as a proxy for 3D structure. Our model, BARNACLE, combines the utilization of unlabeled data through self-supervised pre-training and efficient use of the sparse labeled data through an XGBoost classifier. BARNACLE shows a considerable improvement over both the established classical baseline and a deep neural network. In order to demonstrate that our approach can be applied to tasks with similar data constraints, we show that our findings generalize to the related setting of accessible surface area prediction.

In this paper we discuss the performance of solvers for low-rank linear systems in the tensor-train format, also known as matrix-product states (MPS) in physics. We focus on today's many-core CPU systems and the interplay of the performance and the required linear algebra operations in this setting. Specifically, we consider the tensor-train GMRES method, the modified alternating linear scheme (MALS) and the alternating minimal energy (AMEn). Based on the example of a simple non-symmetric system from discretizing a multidimensional convection-diffusion equation in, e.g., $50^{10}$ grid points, we illustrate the computational complexity of the three methods. This shows that the projection to smaller sub-problems reduces the required number of floating point operations by orders of magnitude: between GMRES and MALS, and again between MALS and AMEn. All three methods require similar underlying linear algebra operations (building blocks). We suggest several building block improvements regarding orthogonalization steps, singular value decompositions, and tensor contractions. In addition, we propose a simple generic preconditioner in the tensor-train format based on a rank-1 approximation of the operator. Combining all optimizations, we obtain roughly a 5x speedup over the reference implementation for the fastest method (AMEn) on a current multi-core CPU.

Developments in numerical simulation of flows and high-performance computing influence one another. More detailed simulation methods create a permanent need for more computational power, while new hardware developments often require changes to the software to exploit new hardware features. This dependency is very pronounced in the case of vector-units which are featured by all modern processors to increase their numerical throughput but require vectorization of the software to be used efficiently. We study the vectorization of a simulation method that exhibits an inherent level of vector-parallelism. This is of particular interest as SIMD operations will hopefully be available with std::simd in a future C++ standard. The simulation method considered here results in the simultaneous solution of multiple sparse linear systems of equations which only differ by their main diagonal and right-hand sides. Such structure arises in the simulation of unsteady flow in turbomachinery by means of a frequency domain approach called harmonic balance.

Natural disasters and epidemics are unfortunate recurring events that lead to huge societal and economic loss. Recent advances in supercomputing can facilitate simulations of such scenarios in (or even ahead of) real-time, therefore supporting the design of adequate responses by public authorities. By incorporating high-velocity data from sensors and modern high-performance computing systems, ensembles of simulations and advanced analysis enable urgent decision-makers to better monitor the disaster and to employ necessary actions (e.g., to evacuate populated areas) for mitigating these events. Unfortunately, frameworks to support such versatile and complex workflows for urgent decision-making are only rarely available and often lack in functionalities. This paper gives an overview of the VESTEC project and framework, which unifies orchestration, simulation, in-situ data analysis, and visualization of natural disasters that can be driven by external sensor data or interactive intervention by the user. We show how different components interact and work together in VESTEC and describe implementation details. To disseminate our experience three different types of disasters are evaluated: a Wildfire in La Jonquera (Spain), a Mosquito-Borne disease in two regions of Italy, and the magnetic reconnection in the Earth magnetosphere.

There are several factorizations of multi-dimensional tensors into lower-dimensional components, known as `tensor networks'. We consider the popular `tensor-train' (TT) format and ask: How efficiently can we compute a low-rank approximation from a full tensor on current multi-core CPUs? Compared to sparse and dense linear algebra, kernel libraries for multi-linear algebra are rare and typically not as well optimized. Linear algebra libraries like BLAS and LAPACK may provide the required operations in principle, but often at the cost of additional data movements for rearranging memory layouts. Furthermore, these libraries are typically optimized for the compute-bound case (e.g.\ square matrix operations) whereas low-rank tensor decompositions lead to memory bandwidth limited operations. We propose a `tensor-train singular value decomposition' (TT-SVD) algorithm based on two building blocks: a `Q-less tall-skinny QR' factorization, and a fused tall-skinny matrix-matrix multiplication and reshape operation. We analyze the performance of the resulting TT-SVD algorithm using the Roofline performance model. In addition, we present performance results for different algorithmic variants for shared-memory as well as distributed-memory architectures. Our experiments show that commonly used TT-SVD implementations suffer severe performance penalties. We conclude that a dedicated library for tensor factorization kernels would benefit the community: Computing a low-rank approximation can be as cheap as reading the data twice from main memory. As a consequence, an implementation that achieves realistic performance will move the limit at which one has to resort to randomized methods that only process part of the data.

We analyze the relaxation of non-pharmaceutical interventions (NPIs) under an increasing number of vaccinations in Germany. For the spread of SARS-CoV-2 we employ a SIR-type model that accounts for age-dependence and includes realistic contact patterns between age groups. The implementation of NPIs occurs on changed contact patterns, improved isolation, or reduced infectiousness when, e.g., wearing masks. We account for spatial heterogeneity and commuting activities in between regions in Germany, and the testing of commuters is considered as a further NPI. We include the ongoing vaccination process and analyze the effect of the B.1.617.2 (Delta) variant, which is considered to be 40%-60% more infectious then the currently dominant B.1.1.7 (Alpha) variant. We explore different opening scenarios under the ongoing vaccination process by assuming that local restrictions are either lifted in early July or August with or without continued wearing of masks and testing. Our results indicate that we can counteract the resurgence of SARS-CoV-2 despite the Delta variant with appropriate timing for the relaxation of NPIs. In all cases, however, school children are hit the hardest.

Background Despite the vaccination process in Germany, a large share of the population is still susceptible to SARS-CoV-2. In addition, we face the spread of novel variants. Until we overcome the pandemic, reasonable mitigation and opening strategies are crucial to balance public health and economic interests. Methods We model the spread of SARS-CoV-2 over the German counties by a graph-SIR-type, metapopulation model with particular focus on commuter testing. We account for political interventions by varying contact reduction values in private and public locations such as homes, schools, workplaces, and other. We consider different levels of lockdown strictness, commuter testing strategies, or the delay of intervention implementation. We conduct numerical simulations to assess the effectiveness of the different intervention strategies after one month. The virus dynamics in the regions (German counties) are initialized randomly with incidences between 75 and 150 weekly new cases per 100,000 inhabitants (red zones) or below (green zones) and consider 25 different initial scenarios of randomly distributed red zones (between 2 and 20% of all counties). To account for uncertainty, we consider an ensemble set of 500 Monte Carlo runs for each scenario. Results We find that the strength of the lockdown in regions with out of control virus dynamics is most important to avoid the spread into neighboring regions. With very strict lockdowns in red zones, commuter testing rates of twice a week can substantially contribute to the safety of adjacent regions. In contrast, the negative effect of less strict interventions can be overcome by high commuter testing rates. A further key contributor is the potential delay of the intervention implementation. In order to keep the spread of the virus under control, strict regional lockdowns with minimum delay and commuter testing of at least twice a week are advisable. If less strict interventions are in favor, substantially increased testing rates are needed to avoid overall higher infection dynamics. Conclusions Our results indicate that local containment of outbreaks and maintenance of low overall incidence is possible even in densely populated and highly connected regions such as Germany or Western Europe. While we demonstrate this on data from Germany, similar patterns of mobility likely exist in many countries and our results are, hence, generalizable to a certain extent.

We discuss some specific software engineering challenges in the field of high-performance computing, and argue that the slow adoption of SE tools and techniques is at least in part caused by the fact that these do not address the HPC challenges `out-of-the-box'. By giving some examples of solutions for designing, testing and benchmarking HPC software, we intend to bring software engineering and HPC closer together.

There are several factorizations of multi-dimensional tensors into lower-dimensional components, known as `tensor networks'. We consider the popular `tensor-train' (TT) format and ask, how efficiently can we compute a low-rank approximation from a full tensor on current multi-core CPUs. Compared to sparse and dense linear algebra, there are much fewer and less extensive well-optimized kernel libraries for multi-linear algebra. Linear algebra libraries like BLAS and LAPACK may provide the required operations in principle, but often at the cost of additional data movements for rearranging memory layouts. Furthermore, these libraries are typically optimized for the compute-bound case (e.g. square matrix operations) whereas low-rank tensor decompositions lead to memory bandwidth limited operations. We propose a `tensor-train singular value decomposition' (TT-SVD) algorithm based on two building blocks: a `Q-less tall-skinny QR' factorization, and a fused tall-skinny matrix-matrix multiplication and reshape operation. We analyze the performance of the resulting TT-SVD algorithm using the Roofline performance model. In addition, we present performance results for different algorithmic variants for shared-memory as well as distributed-memory architectures. Our experiments show that commonly used TT-SVD implementations suffer severe performance penalties. We conclude that a dedicated library for tensor factorization kernels would benefit the community: Computing a low-rank approximation can be as cheap as reading the data twice from main memory. As a consequence, an implementation that achieves realistic performance will move the limit at which one has to resort to randomized methods that only process part of the data.