Quality control (QC) is essential for ensuring the reliability of automated tumor segmentations in breast magnetic resonance imaging (MRI), which are increasingly used in both clinical and research workflows. Manual QC by experts is subjective, time-consuming, and unscalable, highlighting the need for automated solutions. In this work, we propose a 2D convolutional neural network (CNN) with attention-based slice aggregation to classify the quality of tumor segmentations in breast MRI. Leveraging the large-scale, expert-annotated MAMA-MIA dataset, our method processes 2D slices as 2-channel inputs (image and segmentation) and learns to weight their contribution using an attention mechanism for volume-level classification. Our best-performing configuration (attention + augmentation) achieved 64% accuracy, 0.58 F1, and 0.67 AUC, outperforming other settings. These findings show that augmentation and attention are complementary—augmentation increases slice-level diversity, while attention helps identify informative slices among mostly uninformative ones. While performance was moderate, the results establish a baseline for scalable QC in breast MRI, with considerable potential for improvement through enhanced architectures, richer augmentations, and integration with 3D context.

ƒuncX is a distributed function as a service (FaaS) platform that enables flexible, scalable, and high performance remote function execution. Unlike centralized FaaS systems, ƒuncX decouples the cloud-hosted management functionality from the edge-hosted execution functionality. ƒuncX's endpoint software can be deployed, by users or administrators, on arbitrary laptops, clouds, clusters, and supercomputers, in effect turning them into function serving systems. ƒuncX's cloud-hosted service provides a single location for registering, sharing, and managing both functions and endpoints. It allows for transparent, secure, and reliable function execution across the federated ecosystem of endpoints—enabling users to route functions to endpoints based on specific needs. ƒuncX uses containers (e.g., Docker, Singularity, and Shifter) to provide common execution environments across endpoints. ƒuncX implements various container management strategies to execute functions with high performance and efficiency on diverse ƒuncX endpoints. ƒuncX also integrates with an in-memory data store and Globus for managing data that may span endpoints. We motivate the need for ƒuncX, present our prototype design and implementation, and demonstrate, via experiments on two supercomputers, that ƒuncX can scale to more than 130000 concurrent workers. We show that ƒuncX's container warming-aware routing algorithm can reduce the completion time for 3,000 functions by up to 61% compared to a randomized algorithm and the in-memory data store can speed up data transfers by up to 3x compared to a shared file system.

Exploding data volumes and velocities, new computational methods and platforms, and ubiquitous connectivity demand new approaches to computation in the sciences. These new approaches must enable computation to be mobile, so that, for example, it can occur near data, be triggered by events (e.g., arrival of new data), be offloaded to specialized accelerators, or run remotely where resources are available. They also require new design approaches in which monolithic applications can be decomposed into smaller components, that may in turn be executed separately and on the most suitable resources. To address these needs we present funcX---a distributed function as a service (FaaS) platform that enables flexible, scalable, and high performance remote function execution. funcX's endpoint software can transform existing clouds, clusters, and supercomputers into function serving systems, while funcX's cloud-hosted service provides transparent, secure, and reliable function execution across a federated ecosystem of endpoints. We motivate the need for funcX with several scientific case studies, present our prototype design and implementation, show optimizations that deliver throughput in excess of 1 million functions per second, and demonstrate, via experiments on two supercomputers, that funcX can scale to more than more than 130 000 concurrent workers.

The variety of instance types available on cloud platforms offers enormous flexibility to match the requirements of applications with available resources. However, selecting the most suitable instance type and configuring an application to optimally execute on that instance type can be complicated and time-consuming. For example, application parallelism flags must match available cores and problem sizes must be tuned to match available memory. As the search space of application configurations can be enormous, we propose an automated approach, called ParaOpt, to automatically explore and tune application configurations on arbitrary cloud instances. ParaOpt supports arbitrary applications, enables use of custom optimization methods, and can be configured with different optimization targets such as runtime and cost. We evaluate ParaOpt by optimizing genomics, molecular dynamics, and machine learning applications with four types of optimizers. We show with as few as 15 parameterized executions of an application, representing between 1.2%-26.7% of the search space, that ParaOpt is able to identify the optimal configuration in 32.7% of experiments and a near-optimal configuration in 83.2% of cases. As a result of using near-optimal configurations, ParaOpt reduces overall execution time by up to 85.8% when compared with using the default configuration.

High-level programming languages such as Python are increasingly used to provide intuitive interfaces to libraries written in lower-level languages and for assembling applications from various components. This migration towards orchestration rather than implementation, coupled with the growing need for parallel computing (e.g., due to big data and the end of Moore's law), necessitates rethinking how parallelism is expressed in programs. Here, we present Parsl, a parallel scripting library that augments Python with simple, scalable, and flexible constructs for encoding parallelism. These constructs allow Parsl to construct a dynamic dependency graph of components that it can then execute efficiently on one or many processors. Parsl is designed for scalability, with an extensible set of executors tailored to different use cases, such as low-latency, high-throughput, or extreme-scale execution. We show, via experiments on the Blue Waters supercomputer, that Parsl executors can allow Python scripts to execute components with as little as 5 ms of overhead, scale to more than 250000 workers across more than 8000 nodes, and process upward of 1200 tasks per second. Other Parsl features simplify the construction and execution of composite programs by supporting elastic provisioning and scaling of infrastructure, fault-tolerant execution, and integrated wide-area data management. We show that these capabilities satisfy the needs of many-task, interactive, online, and machine learning applications in fields such as biology, cosmology, and materials science.

While the Machine Learning (ML) landscape is evolving rapidly, there has been a relative lag in the development of the “learning systems” needed to enable broad adoption. Furthermore, few such systems are designed to support the specialized requirements of scientific ML. Here we present the Data and Learning Hub for science (DLHub), a multi-tenant system that provides both model repository and serving capabilities with a focus on science applications. DLHub addresses two significant shortcomings in current systems. First, its self-service model repository allows users to share, publish, verify, reproduce, and reuse models, and addresses concerns related to model reproducibility by packaging and distributing models and all constituent components. Second, it implements scalable and low-latency serving capabilities that can leverage parallel and distributed computing resources to democratize access to published models through a simple web interface. Unlike other model serving frameworks, DLHub can store and serve any Python 3-compatible model or processing function, plus multiple-function pipelines. We show that relative to other model serving systems including TensorFlow Serving, SageMaker, and Clipper, DLHub provides greater capabilities, comparable performance without memoization and batching, and significantly better performance when the latter two techniques can be employed. We also describe early uses of DLHub for scientific applications.

Publishing scientific results without the detailed execution environments describing how the results were collected makes it difficult or even impossible for the reader to reproduce the work. However, the configurations of the execution environments are too complex to be described easily by authors. To solve this problem, we propose a framework facilitating the conduct of reproducible research by tracking, creating, and preserving the comprehensive execution environments with Umbrella. The framework includes a lightweight, persistent and deployable execution environment specification, an execution engine which creates the specified execution environments, and an archiver which archives an execution environment into persistent storage services like Amazon S3 and Open Science Framework (OSF). The execution engine utilizes sandbox techniques like virtual machines (VMs), Linux containers and user-space tracers, to create an execution environment, and allows common dependencies like base OS images to be shared by sandboxes for different applications. We evaluate our framework by utilizing it to reproduce three scientific applications from epidemiology, scene rendering, and high energy physics. We evaluate the time and space overhead of reproducing these applications, and the effectiveness of the chosen archive unit and mounting mechanism for allowing different applications to share dependencies. Our results show that these applications can be reproduced using different sandbox techniques successfully and efficiently, even through the overhead and performance slightly vary.

The high energy physics (HEP) community relies upon a global network of computing and data centers to analyze data produced by multiple experiments at the Large Hadron Collider (LHC). However, this global network does not satisfy all research needs. Ambitious researchers often wish to harness computing resources that are not integrated into the global network, including private clusters, commercial clouds, and other production grids. To enable these use cases, we have constructed Lobster, a system for deploying data intensive high throughput applications on non-dedicated clusters. This requires solving multiple problems related to non-dedicated resources, including work decomposition, software delivery, concurrency management, data access, data merging, and performance troubleshooting. With these techniques, we demonstrate Lobster running effectively on 10k cores, producing throughput at a level comparable with some of the largest dedicated clusters in the LHC infrastructure.

The computing needs of high energy physics experiments like the Compact Muon Solenoid experiment at the Large Hadron Collider currently exceed the available dedicated computational resources, hence motivating a push to leverage opportunistic resources. However, access to opportunistic resources faces many obstacles, not the least of which is making available the complex software stack typically associated with such computations. This paper describes a framework constructed using existing software packages to distribute the needed software to opportunistic resources without the need for the job to have root-level privileges. Preliminary tests with this framework have demonstrated the feasibility of the approach and identified bottlenecks as well as reliability issues which must be resolved in order to make this approach viable for broad use.