Data distribution across different facilities offers benefits such as enhanced resource utilization, increased resilience through replication, and improved performance by processing data near its source. However, managing such data is challenging due to heterogeneous access protocols, disparate authentication models, and the lack of a unified coordination framework. This paper presents DynoStore, a system that manages data across heterogeneous storage systems. At the core of DynoStore are data containers, an abstraction that provides standardized interfaces for seamless data management, irrespective of the underlying storage systems. Multiple data container connections create a cohesive wide-area storage network, ensuring resilience using erasure coding policies. Furthermore, a load-balancing algorithm ensures equitable and efficient utilization of storage resources. We evaluate DynoStore using benchmarks and realworld case studies, including the management of medical and satellite data across geographically distributed environments. Our results demonstrate a 10 % performance improvement compared to centralized cloud-hosted systems while maintaining competitive performance with state-of-the-art solutions such as Redis and IPFS. DynoStore also exhibits superior fault tolerance, withstanding more failures than traditional systems.

Failures in Task-based Parallel Programming (TBPP) can severely degrade performance and result in incomplete or incorrect outcomes. Existing failure-handling approaches, including reactive, proactive, and resilient methods such as retry and checkpointing mechanisms, often apply uniform retry mechanisms regardless of the root cause of failures, failing to account for the unique characteristics of TBPP frameworks such as heterogeneous resource availability and task-level failures. To address these limitations, we propose Wrath, a novel systematic approach that categorizes failures based on the unique layered structure of TBPP frameworks and defines specific responses to address failures at different layers. Wrath combines a distributed monitoring system and a resilient module to collaboratively address different types of failures in real time. The monitoring system captures execution and resource information, reports failures, and profiles tasks across different layers of TBPP frameworks. The resilient module then categorizes failures and responds with appropriate actions, such as hierarchically retrying failed tasks on suitable resources. Evaluations demonstrate that Wrath significantly improves TBPP robustness, tripling the task success rate and maintaining an application success rate of over 90 % for resolvable failures. Additionally, Wrath can reduce the time to failure by$20 \%-50 \%$, allowing tasks that are destined to fail to be identified and fail more quickly.

The exponential growth of data necessitates distributed storage models, such as peer-to-peer systems and data federations.While distributed storage can reduce costs and increase reliability, the heterogeneity in storage capacity, I/O performance, and failure rates of storage resources makes their efficient use a challenge.Further, node failures are common and can lead to data unavailability and even data loss.

High-performance computing (HPC) systems and cloud data centers are converging, and containers are becoming the default method of portable software deployment. Yet, while containers simplify software management, they face significant performance challenges in HPC environments as they must sacrifice hardware-specific optimizations to achieve portability. Although HPC containers can use runtime hooks to access optimized MPI libraries and GPU devices, they are limited by application binary interface (ABI) compatibility and cannot overcome the effects of early-stage compilation decisions. Acceleration as a Service (XaaS) proposes a vision of performance-portable containers, where a containerized application should achieve peak performance across all HPC systems. We present a practical realization of this vision through Source and Intermediate Representation (IR) containers, where we delay performance-critical decisions until the target system specification is known. We analyze specialization mechanisms in HPC software and propose a new LLM-assisted method for automatic discovery of specializations. By examining the compilation pipeline, we develop a methodology to build containers optimized for target architectures at deployment time. Our prototype demonstrates that new XaaS containers combine the convenience of containerization with the performance benefits of system-specialized builds.

The high-performance computing (HPC) community has adopted incentive structures to motivate reproducible research, with major conferences awarding badges to papers that meet reproducibility requirements. Yet, many papers do not meet such requirements. The uniqueness of HPC infrastructure and software, coupled with strict access requirements, may limit opportunities for reproducibility. In the absence of resource access, we believe that regular documented testing, through continuous integration (CI), coupled with complete provenance information, can be used as a substitute. Here, we argue that better HPC-compliant CI solutions will improve reproducibility of applications. We present a survey of reproducibility initiatives and describe the barriers to reproducibility in HPC. To address existing limitations, we present a GitHub Action, CORRECT, that enables secure execution of tests on remote HPC resources. We evaluate CORRECT’s usability across three different types of HPC applications, demonstrating the effectiveness of using CORRECT for automating and documenting reproducibility evaluations.

Efforts to reduce the environmental impact of HPC often focus on resource providers, but choices made by users, e.g., concerning where to run, can be equally consequential. Here we present evidence that new accounting methods that charge users for energy used can incentivize significantly more efficient behavior. We first survey 300 HPC users and find that fewer than 30% are aware of their energy consumption, and that energy efficiency is a low priority concern. We then propose two new multi-resource accounting methods that charge for computations based on their energy consumption or carbon footprint, respectively. Finally, we conduct both simulation studies and a user study to evaluate the impact of these two methods on user behavior. We find that while only providing users feedback on their energy use had no impact on their behavior, associating energy with cost incentivized users to select more efficient resources, and use 40% less energy.

Workflow and serverless frameworks have empowered new approaches to distributed application design by abstracting compute resources. However, their typically limited or one-size-fits-all support for advanced data flow patterns leaves optimization to the application programmer—optimization that becomes more difficult as data become larger. The transparent object proxy, which provides wide-area references that can resolve to data regardless of location, has been demonstrated as an effective low-level building block in such situations. Here we propose three high-level proxy-based programming patterns—distributed futures, streaming, and ownership—that make the power of the proxy pattern usable for more complex and dynamic distributed program structures. We motivate these patterns via careful review of application requirements and describe implementations of each pattern. We evaluate our implementations through a suite of benchmarks and by applying them in three meaningful scientific applications, in which we demonstrate substantial improvements in runtime, throughput, and memory usage.

The need for real-time processing to enable automated decision making and experimental steering has driven a shift from high-performance computing workflows on a centralized system to a distributed approach that integrates remote data sources, edge devices, and diverse compute facilities. Under this paradigm, data can be processed close to the source where it is generated, thus reducing latency and bandwidth usage. System resilience is thus a key challenge, requiring distributed workflows to survive component failures and to meet stringent quality-of-service requirements, which results in the need to mitigate anomalies such as congestion and low availability of resources. To address these challenges, we propose Diaspora, a unified resilience framework that is inspired by event-driven communication patterns used in public clouds. Specifically, we propose an event fabric that extends across sites, facilities, and computations to provide timely, reliable, and accurate information about data, application, and resource status. On top of the event fabric, we build resilience-enabling services that combine QoS-aware data streaming, resilient data views, resilient compute and data resources, and anomaly detection and prediction, all of which collectively enhance workflow resilience for these scientific cases.

Task-based execution frameworks, such as parallel programming libraries, computational workflow systems, and function-as-a-service platforms, enable the composition of distinct tasks into a single, unified application designed to achieve a computational goal and abstract the parallel and distributed execution of those tasks on arbitrary hardware. Research into these task executors has accelerated as computational sciences increasingly need to take advantage of parallel compute and/or heterogeneous hardware. However, the lack of evaluation standards makes it challenging to compare and contrast novel systems against existing implementations. Here, we introduce TaPS, the Task Performance Suite, to support continued research in distributed task executor frameworks. TaPS provides (1) a unified, modular interface for writing and evaluating applications using arbitrary execution frameworks and data management systems and (2) an initial set of reference synthetic and real-world science applications. We discuss how the design of TaPS supports the reliable evaluation of frameworks and demonstrate TaPS through a survey of benchmarks using the provided reference applications.

Python has become the language of choice for managing many scientific applications. However, when distributing a Python application, it is necessary that all application dependencies be distributed and available in the target execution environment. A specific consequence is that Python workflows suffer from slow scale out due to the time required to import dependencies. We describe ProxyImports, a method to package and distribute Python dependencies in a lazy fashion while remaining transparent and easy to use. Using ProxyImports, Python packages are loaded only once (e.g., by a workflow head node) and are transferred asynchronously to compute nodes. We evaluate our implementation on the Perlmutter and Theta supercomputers and in an HPC cloud-bursting scenario. Our experiments show that ProxyImports significantly reduces the average time to import large modules across an HPC system and demonstrate that this method can be used easily to distribute user-packages to cloud resources. We conclude that ProxyImports improves application runtime, reduces contention on metadata servers and facilitates runtime portability of Python applications.

Advances in networks, accelerators, and cloud services encourage programmers to reconsider where to compute---such as when fast networks make it cost-effective to compute on remote accelerators despite added latency. Workflow and cloud-hosted serverless computing frameworks can manage multi-step computations spanning federated collections of cloud, high-performance computing (HPC), and edge systems, but passing data among computational steps via cloud storage can incur high costs. Here, we overcome this obstacle with a new programming paradigm that decouples control flow from data flow by extending the pass-by-reference model to distributed applications. We describe ProxyStore, a system that implements this paradigm by providing object proxies that act as wide-area object references with just-in-time resolution. This proxy model enables data producers to communicate data unilaterally, transparently, and efficiently to both local and remote consumers. We demonstrate the benefits of this model with synthetic benchmarks and real-world scientific applications, running across various computing platforms.

Summary The general increase in data size and data sharing motivates the adoption of Big Data strategies in several scientific disciplines. However, while several options are available, no particular guidelines exist for selecting a Big Data engine. In this paper, we compare the runtime performance of two popular Big Data engines with Python APIs, Apache Spark, and Dask, in processing neuroimaging pipelines. Our experiments use three synthetic neuroimaging applications to process the 606 GB BigBrain image and an actual pipeline to process data from thousands of anatomical images. We benchmark these applications on a dedicated HPC cluster running the Lustre file system while using varying combinations of the number of nodes, file size, and task duration. Our results show that although there are slight differences between Dask and Spark, the performance of the engines is comparable for data‐intensive applications. However, Spark requires more memory than Dask, which can lead to slower runtime depending on configuration and infrastructure. In general, the limiting factor was the data transfer time. While both engines are suitable for neuroimaging, more efforts need to be put to reduce the data transfer time and the memory footprint of applications.

The emergence of Big Data in recent years has resulted in a growing need for efficient data processing solutions. While infrastructures with sufficient compute power are available, the I/O bottleneck remains. The Linux page cache is an efficient approach to reduce I/O overheads, but few experimental studies of its interactions with Big Data applications exist, partly due to limitations of real-world experiments. Simulation is a popular approach to address these issues, however, existing simulation frameworks do not simulate page caching fully, or even at all. As a result, simulation-based performance studies of data-intensive applications can lead to misleading results and inaccurate conclusions.In this paper, we propose an I/O simulation model that captures the key features of the Linux page cache. We have implemented this model as part of the WRENCH workflow simulation framework, which itself builds on the popular Sim-Grid distributed systems simulation framework. Our model and its implementation enable the simulation of both single-threaded and multithreaded applications, and of both writeback and writethrough caches for local or network-based filesystems. We evaluate the accuracy of our model in different conditions, including sequential and concurrent applications, as well as local and remote I/Os. We find that our page cache model reduces the simulation error by up to an order of magnitude when compared to state-of-the-art, cacheless simulations. Our model is publicly available in the WRENCH framework, making it usable in a wide range of simulation studies.

Open-access neuroimaging datasets have reached petabyte scale and continue to grow. The ability to leverage the entirety of these datasets is limited to a restricted number of research laboratories with both the capacity and infrastructure to process the data. Whereas Big Data engines have significantly reduced application performance penalties with respect to data movement, their applied strategies (e.g. data locality, in-memory computing and lazy evaluation) are not necessarily practical within neuroimaging workflows where intermediary results may need to be materialized to shared storage for post-processing analysis. In this paper we evaluate the performance advantage brought by Intel®Optane™ DC persistent memory for the processing of large neuroimaging datasets using the two available configurations modes: Memory mode and App Direct mode. We employ a synthetic algorithm on the 76 GiB and 603 GiB BigBrain, as well as apply a standard neuroimaging application on the Consortium for Reliability and Reproducibility (CoRR) dataset using 25 and 96 parallel processes in both cases. Our results show that the performance of applications leveraging persistent memory is superior to that of other storage devices, with the exception of DRAM. This is the case in both Memory and App Direct mode, irrespective of the amount of data and parallelism. Furthermore, persistent memory in App Direct mode is believed to benefit from the use of DRAM as a cache for writing when output data is significantly smaller than available memory. We believe the use of persistent memory will be beneficial to both neuroimaging applications running on HPC or visualization of large, high-resolution images.

Neuroimaging datasets are rapidly growing in size as a result of advancements in image acquisition methods, open-science and data sharing. However, the adoption of Big Data processing strategies by neuroimaging processing engines remains limited. Here, we evaluate three Big Data processing strategies (in-memory computing, data locality and lazy evaluation) on typical neuroimaging use cases, represented by the BigBrain dataset. We contrast these various strategies using Apache Spark and Nipype as our representative Big Data and neuroimaging processing engines, on Dell EMC's Top-500 cluster. Big Data thresholds were modeled by comparing the data-write rate of the application to the filesystem bandwidth and number of concurrent processes. This model acknowledges the fact that page caching provided by the Linux kernel is critical to the performance of Big Data applications. Results show that in-memory computing alone speeds-up executions by a factor of up to 1.6, whereas when combined with data locality, this factor reaches 5.3. Lazy evaluation strategies were found to increase the likelihood of cache hits, further improving processing time. Such important speed-up values are likely to be observed on typical image processing operations performed on images of size larger than 75GB. A ballpark speculation from our model showed that in-memory computing alone will not speed-up current functional MRI analyses unless coupled with data locality and processing around 280 subjects concurrently. Furthermore, we observe that emulating in-memory computing using in-memory file systems (tmpfs) does not reach the performance of an in-memory engine, presumably due to swapping to disk and the lack of data cleanup. We conclude that Big Data processing strategies are worth developing for neuroimaging applications.

Big Data has become prominent throughout many scientific fields, and as a result, scientific communities have sought out Big Data frameworks to accelerate the processing of their increasingly data-intensive pipelines. However, while scientific communities typically rely on High-Performance Computing (HPC) clusters for the parallelization of their pipelines, many popular Big Data frameworks such as Hadoop and Apache Spark were primarily designed to be executed on dedicated commodity infrastructures. This paper evaluates the benefits of pilot jobs over traditional batch submission for Apache Spark on HPC clusters. Surprisingly, our results show that the speed-up provided by pilot jobs over batch scheduling is moderate to non-existent (0.98 on average) despite the presence of long queuing times. In addition, pilot jobs provide an extra layer of scheduling that complicates debugging and deployment. We conclude that traditional batch scheduling should remain the default strategy to deploy Apache Spark applications on HPC clusters.

Splitting and merging data is a requirement for many parallel or distributed processing operations. Naive algorithms to split and merge 3D blocks from ultra-high resolution images perform very poorly, as a result of seek times. In contrast, naive algorithms to split and merge 3D slabs perform optimally as seek time is significantly minimized. We introduce and analyze sequential algorithms (Clustered reads, Multiple reads, Clustered writes, and Multiple writes) that leverage memory buffering to address this issue. Clustered reads and Clustered writes, access image chunks only once, but they have to seek in the reconstructed image. Multiple reads and Multiple writes minimize seeks in the reconstructed image, but they access image chunks multiple times. Evaluation on a 3850×3025×3500 brain image shows that our algorithms perform similarly to the optimal configuration provided that enough memory is available. Additionally, Multiple reads supports on-the-fly compression of the merged image transparently but Clustered reads does not, due to its use of negative seeking. We conclude that splitting and merging large 3D images can be done efficiently without relying on complex data formats.