In modern science, the growing complexity of large-scale scientific projects has led to an increasing reliance on cross-facility scientific workflows, where resources and expertise from multiple institutions and geographic locations are leveraged to accelerate scientific discovery. These workflows often require transmitting huge amounts of scientific data through wide-area networks. Although high-speed networks like ESnet and transfer services such as Globus have improved data mobility, several challenges remain. The sheer volume of data can overwhelm network bandwidth, widely used transport protocols such as TCP suffer from inefficiencies due to retransmissions triggered by packet loss, and existing fault-tolerance mechanisms like erasure coding introduce substantial overhead.

Scientific applications generate an unprecedented volume of data, overwhelming the network and file systems’ bandwidth and posing challenges for efficient and scalable data retrieval and analysis. Progressive data compression offers a promising solution by enabling on-demand retrieval at reduced size. However, existing progressive methods either fail to bound the errors in essential quantities of interest (QoIs) derived from raw data or suffer from suboptimal retrieval efficiency. In this work, we propose QProR, an efficient QoI-based progressive framework that optimizes progressive retrieval for target QoIs. Our key contributions include: (1) a systematic framework that integrates error-controlled lossy compressors with bitplane encoding while decoupling the two processes for high flexibility and adaptability; (2) a novel weighted bitplane encoding method which incorperates QoI knowledge into data refactoring to enhance retrieval efficiency; (3) an optimized retrieval strategy that accounts for the varying impacts of different variables on multivariate QoIs; (4) comprehensive evaluations using six real-world datasets from multiple scientific applications and thorough comparisons against state of the arts. Experimental results demonstrate that QProR achieves up to \(80.38\%\) reduction in the retrieval size under the same requested QoI error tolerance, when compared with the best-performing existing methods. When transferring 384 GB of scientific data to remote sites, QProR delivers up to 1.68 × speedup in the end-to-end data transfer performance.

This paper presents a novel approach to the joint optimization of job scheduling and data allocation in grid computing environments. We formulate this joint optimization problem as a mixed integer quadratically constrained program. To tackle the nonlinearity in the constraint, we alternatively fix a subset of decision variables and optimize the remaining ones via Mixed Integer Linear Programming (MILP). We solve the MILP problem at each iteration via an off-the-shelf MILP solver. Our experimental results show that our method significantly outperforms existing heuristic methods, employing either independent optimization or joint optimization strategies. We have also verified the generalization ability of our method over grid environments with various sizes and its high robustness to the algorithm setting.

Cache systems are widely used to speed up data retrieving. Modern HPC, data analytics, and AI/ML workloads generate vast, multi-dimensional datasets, and those data are accessed via complex queries. However, the probability of requesting the exact same data across different queries is low, leading to limited performance improvement when a traditional key-value cache is applied. In this paper, we present Mosaic-Cache, a proactive and general caching framework that enables applications with efficient partial overlapped data reuse through novel overlap-aware cache interfaces for fast content-level reuse. The core components include a metadata manager leveraging customizable indexing for fast overlap lookups, an adaptive fetch planner for dynamic cache-to-storage decisions, and an async merger to reduce cache fragmentation and redundancy. Evaluations on real-world HPC datasets show that Mosaic-Cache improves overall performance by up to 4.1× over traditional key-value-based cache while adding minimal overhead in worst-case scenarios.

Neural networks are increasingly integrated into scientific discovery, where input data reduction and model quantization play a key role in accelerating inference. However, understanding and mitigating the impact of these techniques on output error is critical for ensuring reliable results, particularly in tasks demanding high numerical precision. This paper introduces a comprehensive framework for optimizing neural network inference in scientific computing by combining data reduction and model quantization while maintaining error-controlled outcomes. We develop theoretical analyses to bound error propagation under these techniques and propose a framework that balances computational performance with error constraints. Evaluation on real-world learning-based combustion simulations and satellite image classification shows that our derived error bounds accurately predict observed errors while enabling significant computational speedup under our framework. This work highlights the potential for further leveraging advancements in modern lossy compression algorithms and hardware accelerators that support lower-precision formats.

The rapid growth in scientific data generation is outpacing advancements in computing systems necessary for efficient storage, transfer, and analysis, particularly in the context of exascale computing. With the deployment of first-generation exascale computing systems and next-generation experimental facilities, this gap is widening and necessitates effective data reduction techniques to manage enormous data volumes. Over the past decade, various data reduction methods, including lossless compression, error-controlled lossy compression, and data refactoring, have been developed to accelerate I/O in scientific workflows. Despite significant reductions in data volume, these methods introduce considerable computational overhead, which can become the new bottleneck in data processing. To mitigate this, GPU-accelerated data reduction algorithms have been introduced. However, challenges remain in their integration into exascale workflows, including limited portability across different GPU architectures, substantial memory transfer overhead, and reduced scalability on dense multi-GPU systems. To address these challenges, we propose HPDR, a high-performance and portable data reduction framework. HPDR is designed to enable the execution of state-of-the-art reduction algorithms across diverse processor architectures while reducing memory transfer overhead to 2.3 % of the original, resulting in up to$3.5 \times$faster throughput compared to existing solutions. It also achieves up to 96% of the theoretical speedup in multi-GPU settings. In addition, evaluations on accelerating I/O operations at scale up to 1,024 nodes of the Frontier supercomputer demonstrate that HPDR can achieve up to$103\ \text{TB} / \mathrm{s}$reduction throughput, providing up to$4 \times$acceleration in parallel I/O performance compared to existing data reduction routines. This work highlights the potential of HPDR to significantly enhance data reduction efficiency in exascale computing environments.

Large-scale simulations or scientific experiments produce petabytes of data per run. This poses massive challenges for I/O and storage when scientific analysis workflows are run manually offline. Unsupervised deep learning-based techniques to extract patterns and non-linear relations from these large amounts of data provide a way to build scientific understanding from raw data, reducing the need for manual pre-selection of analysis steps, but require exascale compute and memory to process the full dataset available. In this paper, we demonstrate a heterogeneous streaming workflow in which plasma simulation data is streamed directly to a Machine Learning (ML) application training a model on the simulation data in-transit, completely circumventing the capacity-constrained filesystem bottleneck. This workflow employs openPMD to provide a high level interface to describe scientific data and also uses ADIOS2, to transfer volumes of data that exceed the capabilities of the filesystem. We employ experience replay to avoid catastrophic forgetting in learning from this non-steady state process in a continual manner and adapt it to improve model convergence while learning in-transit. As a proof-of-concept, we approach the ill-posed inverse problem of predicting particle dynamics from radiation in a particle-incell (PIConGPU) simulation of the Kelvin-Helmholtz instability (KHI). We detail hardware-software co-design challenges as we scale PIConGPU to full Frontier, the Top-1 system as of June 2024 Top500 list.

Scientific applications produce vast amounts of data, posing grand challenges in the underlying data management and analytic tasks. Progressive compression is a promising way to address this problem, as it allows for on-demand data retrieval with significantly reduced data movement cost. However, most existing progressive methods are designed for CPUs, leaving a gap for them to unleash the power of today’s heterogeneous computing systems with GPUs.In this work, we propose HP-MDR, a high-performance and portable data refactoring and progressive retrieval framework for GPUs. Our contributions are four-fold: (1) We carefully optimize the bitplane encoding and lossless encoding, two key stages in progressive methods, to achieve high performance on GPUs; (2) We propose pipeline optimization and incorporate it with data refactoring and progressive retrieval workflows to further enhance the performance for large data process; (3) We leverage our framework to enable high-performance data retrieval with guaranteed error control for common Quantities of Interest; (4) We evaluate HP-MDR and compare it with state of the arts using five real-world datasets. Experimental results demonstrate that HP-MDR delivers an average 13.68 × and 6.31 × throughput in data refactoring and progressive retrieval tasks, respectively. It also leads to 11.22 × throughput for recomposing required data representations under Quantity-of-Interest error control and 6.04 × performance for the corresponding end-to-end data retrieval, when compared with state-of-the-art solutions.

We present a general framework for compressing unstructured scientific data with known local connectivity. A common application is simulation data defined on arbitrary finite element meshes. The framework employs a greedy topology preserving reordering of original nodes which allows for seamless integration into existing data processing pipelines. This reordering process depends solely on mesh connectivity and can be performed offline for optimal efficiency. However, the algorithm’s greedy nature also supports on-the-fly implementation. The proposed method is compatible with any compression algorithm that leverages spatial correlations within the data. The effectiveness of this approach is demonstrated on a large-scale real dataset using several compression methods, including MGARD, SZ, and ZFP.

Data compression plays a key role in reducing storage and I/O costs. Traditional lossy methods primarily target data on rectilinear grids and cannot leverage the spatial coherence in unstructured mesh data, leading to suboptimal compression ratios. We present a multi-component, error-bounded compression framework designed to enhance the compression of floating-point unstructured mesh data, which is common in scientific applications. Our approach involves interpolating mesh data onto a rectilinear grid and then separately compressing the grid interpolation and the interpolation residuals. This method is general, independent of mesh types and typologies, and can be seamlessly integrated with existing lossy compressors for improved performance. We evaluated our framework across twelve variables from two synthetic datasets and two real-world simulation datasets. The results indicate that the multi-component framework consistently outperforms state-of-the-art lossy compressors on unstructured data, achieving, on average, a 2.3 − 3.5× improvement in compression ratios, with error bounds ranging from 1 × 10 the−6to 1×10−2. We further investigate impact of hyperparameters, such as grid spacing and error allocation, to deliver optimal compression ratios in diverse datasets.

Nuclear fusion holds the promise of an endless source of energy. Several research experiments across the world and joint modeling and simulation efforts between the nuclear physics and high performance computing communities are actively preparing the operation of the International Thermonuclear Experimental Reactor (ITER). Both experimental reactors and their simulated counterparts generate data that must be analyzed quickly and in a resilient way to support decision making for the configuration of subsequent runs or prevent a catastrophic failure. However, the cost if the traditional techniques used to improve the resilience of analysis workflows, i.e., replicating datasets and computational tasks, becomes prohibitive with explosion of the volume of data produced by modern instruments and simulations. Therefore, we advocate in this paper for an alternate approach based on data reduction and data streaming. The rationale is that by allowing for a reasonable, controlled, and guaranteed loss of accuracy it becomes possible to transfer smaller amounts of data, shorten the execution time of analysis workflows, and lower the cost of replication to increase resilience. We develop our research and development roadmap towards resilient near real-time analysis workflows in fusion energy science and present early results showing that data streaming and data reduction is a promising way to speed up the execution and improve the resilience of analysis workflows.

The ever-increasing volume of data produced by HPC simulations necessitates scalable methods for data exploration and knowledge extraction. Scientific data analysis often involves complex queries across distributed datasets, requiring manipulation of multiple primary variables and generating derived data that needs to be handled efficiently, creating challenges for applications that need to parse many large datasets. Relying on individual applications to handle all intermediate data generally leads to redundant computations across studies and unnecessary data transfers. In this paper, we investigate the performance of different approaches where applications define derived variables as quantities of interest (QoIs) and offload the computation and transfer of these QoIs to the I/O library. This significantly reduces redundancy and optimizes data movement across the distributed storage and processing infrastructure by allowing control over when and where derived variables are computed. We present a detailed analysis of the performance-storage trade-offs associated with different solutions and showcase results for our study on two large-scale datasets created from climate and combustion simulations.

As simulation-based scientific discovery advances to exascale, a major question that the community is striving to answer is how to co-design data storage and complex physicsrich analytics in a way that the time to knowledge can be minimized for post-processing. A particular challenge is how to accommodate a broad spectrum of data analytics needsparticularly those that become clear only until very late during the post-processing, a scenario where existing methods, such as in situ processing, are unable or less effective in supporting data analytics. As HPC storage systems have become deeper and more complex with the recent addition of NVMe, die-stacked memory, and burst buffer, it requires fundamentally rethinking new paradigms and methods for data storage and analysis. This paper aims to address the issue of I/O interference for data analytics over local ephemeral storage, which is shared by multiple applications in a non-exclusive node usage scenario-often configured for small- to medium-sized clusters. At the core of this work is a coordinated cross-layer approach that reacts to storage interference from both storage and application layers. By decomposing and distributing analysis data across the storage hierarchy, data analytics can adapt to the interference by reducing or completely avoiding access to lower tiers whenever there is a high interference, while maintaining a prescribed error bound to limit the information loss. Meanwhile, proper actions are also taken at the storage layer to ensure sufficient bandwidth is allocated for retrieving an augmentation, which is based upon the cardinality and accuracy of the augmentation as well as the nature of an application. We evaluate three realworld data analytics, XGC, GenASiS, and CFD, on Chameleon, and quantitatively demonstrate that the I/O performance can be vastly improved, e.g., by 52% versus no adaptivity and 36% versus single-layer adaptivity, while maintaining acceptable outcomes of data analysis.

The unprecedented amount of scientific data has introduced heavy pressure on the current data storage and transmission systems. Progressive compression has been proposed to mitigate this problem, which offers data access with on-demand precision. However, existing approaches only consider precision control on primary data, leaving uncertainties on the quantities of interest (QoIs) derived from it. In this work, we present a progressive data retrieval framework with guaranteed error control on derivable QoIs. Our contributions are three-fold. (1) We carefully derive the theories to strictly control QoI errors during progressive retrieval. Our theory is generic and can be applied to any QoIs that can be composited by the basis of derivable QoIs proved in the paper. (2) We design and develop a generic progressive retrieval framework based on the proposed theories, and optimize it by exploring feasible progressive representations. (3) We evaluate our framework using five real-world datasets with a diverse set of QoIs. Experiments demonstrate that our framework can faithfully respect any user-specified QoI error bounds in the evaluated applications. This leads to over $2.02 \times$ performance gain in data transfer tasks compared to transferring the primary data while guaranteeing a QoI error that is less than 1E-5.

Scientific discoveries are increasingly constrained by limited storage space and I/O capacities. For time-series simulations and experiments, their data often need to be decimated over timesteps to accommodate storage and I/O limitations. In this paper, we propose a technique that addresses storage costs while improving post-analysis accuracy through spatiotemporal adaptive, error-controlled lossy compression. We investigate the trade-off between data precision and temporal output rates, revealing that reducing data precision and increasing timestep frequency lead to more accurate analysis outcomes. Additionally, we integrate spatiotemporal feature detection with data compression and demonstrate that performing adaptive error-bounded compression in higher dimensional space enables greater compression ratios, leveraging the error propagation theory of a transformation-based compressor. To evaluate our approach, we conduct experiments using the well-known E3SM climate simulation code and apply our method to compress variables used for cyclone tracking. Our results show a significant reduction in storage size while enhancing the quality of cyclone tracking analysis, both quantitatively and qualitatively, in comparison to the prevalent timestep decimation approach. Compared to three state-of-the-art lossy compressors lacking feature preservation capabilities, our adaptive compression framework improves perfectly matched cases in TC tracking by 26.4-51.3% at medium compression ratios and by 77.3-571.1% at large compression ratios, with a merely 5–11% computational overhead.

As the disparity between compute and I/O on high-performance computing systems has continued to widen, it has become increasingly difficult to perform post-hoc data analytics on full-resolution scientific simulation data due to the high I/O cost. Error-bounded data decomposition and progressive data retrieval framework has recently been developed to address such a challenge by performing data decomposition before storage and reading only part of the decomposed data when necessary. However, the performance of the progressive retrieval framework has been suffering from the over-pessimistic error control theory, such that the achieved maximum error of recomposed data is significantly lower than the required error. Therefore, more data than required is fetched for recomposition, incurring additional I/O overhead. In order to tackle this issue, we propose a DNN-based progressive retrieval framework that can better identify the minimum amount of data to be retrieved. Our contributions are as follows: 1) We provide an in-depth investigation of the recently developed progressive retrieval framework; 2) We propose two designs of prediction models (named D-MGARD and E-MGARD) to estimate the amount of retrieved data size based on error bounds. 3) We evaluate our proposed solutions using scientific datasets generated by real-world simulations from two domains. Evaluation results demonstrate the effectiveness of our solution in accurately predicting the amount of retrieval data size, as well as the advantages of our solution over the traditional approach to reducing the I/O overhead. Based on our evaluation, our solution is shown to read significantly less data (5% - 40% with D-MGARD, 20% - 80% with E-MGARD).

The current practice for fusion scientists running first principle simulations on high performance computing plat-forms is to either run their simulations and output their data for post-hoc analysis, or to place in situ analytics into their code. In this paper we examine a complex workflow using XGC fusions simulation run on the Oak Ridge Leadership Computing Facility's supercomputer Summit, which also involve three anal-yses as part of the results necessary for scientific discovery. We discuss the challenges faced when implementing these algorithms and present an original hybrid staging technique to help enable the physicists to make discoveries during the execution of the simulation. By creating this infrastructure, we can examine complicated physics results, which may not have been possible without the infrastructure. For example, our work enables the online visualization of turbulent homoclinic tangle around the magnetic X-point, breaking the last confinement surface. This visualization could help fusion scientists to better understand and improve the turbulence spread of plasma exhaust heat, which is crucial toward realizing plasmas beyond the currently accessible physics regimes of present-day tokamak reactors. The physics of turbulent homoclinic tangle will be reported in a future physics publication, by utilizing the original online analysis/visualization framework presented in this paper.

Solid state storage systems have been very effectively used in small devices; however, their effectiveness for large systems such as supercomputers is not yet proven. Recently, for the first time, a new supercomputer is being deployed with an all-flash storage as its main file system. In this work, we report our preliminary study of the I/O performance on this supercomputer named Perlmutter. We are able to achieve 1.4 TB/s with the default file configuration on the system. This default configuration outperforms dozens of other choices tested, though the current observed performance is still pretty far from the theoretical peak performance of 5 TB/s.

Nowadays, data reduction is becoming increasingly important in dealing with the large amounts of scientific data. Existing multilevel compression algorithms offer a promising way to manage scientific data at scale, but may suffer from relatively low performance and reduction quality. In this paper, we propose MGARD+, a multilevel data reduction and refactoring framework drawing on previous multilevel methods, to achieve high-performance data decomposition and high-quality error-bounded lossy compression. Our contributions are four-fold: 1) We propose to leverage a level-wise coefficient quantization method, which uses different error tolerances to quantize the multilevel coefficients. 2) We propose an adaptive decomposition method which treats the multilevel decomposition as a preconditioner and terminates the decomposition process at an appropriate level. 3) We leverage a set of algorithmic optimization strategies to significantly improve the performance of multilevel decomposition/recomposition. 4) We evaluate our proposed method using four real-world scientific datasets and compare with several state-of-the-art lossy compressors. Experiments demonstrate that our optimizations improve the decomposition/recomposition performance of the existing multilevel method by up to$70 \times$, and the proposed compression method can improve compression ratio by up to$2 \times$compared with other state-of-the-art error-bounded lossy compressors under the same level of data distortion.

The rate of data generated by cutting-edge experimental science facilities and large-scale simulations enabled by current high-performance computing (HPC) systems has continued to grow at a far greater pace than the development of the network and storage capabilities on which these systems rely. To cope with this challenge, scientist are moving toward the creation of autonomous experiments and HPC simulations using machine learning. However, efficiently moving, storing, and processing large amounts of data away from the point of origin presents an incredible challenge. In-memory computing, in situ analysis, data staging, and data streaming are recognized viable alternatives to traditional file-based methods for transferring data between coupled workflows. However, the performance trade-offs and limitations for these methods are not fully understood when used in HPC applications. This article presents a comprehensive performance assessment of the current solutions for data staging when applied to applications that are not necessary I/O intensive which makes them not ideal candidates for these methods. Our study is based on experiments running at scale on Oak Ridge National Laboratory's Summit supercomputer using applications and simulations that cover typical computational motifs and patterns. We investigated the usability and cost/benefit trade-offs of staging algorithms for HPC applications under different scenarios and highlight opportunities for optimizing the dataflow between coupled simulation workflows.

Scientific simulations on high-performance computing systems produce vast amounts of data that need to be stored and analyzed efficiently. Lossy compression significantly reduces the data volume by trading accuracy for performance. Despite the recent success of lossy compressions, such as ZFP and SZ, the compression performance is still far from being able to keep up with the exponential growth of data. This article aims to further take advantage of application characteristics, an area that is often under-explored, to improve the compression ratios of adaptive mesh refinement (AMR) - a widely used numerical solver that allows for an improved resolution in limited regions. We propose a level reordering techniquezMeshto reduce the storage footprint of AMR applications. In particular, we group the data points that are mapped to the same or adjacent geometric coordinates such that the dataset is smoother and more compressible. Unlike the prior work where the compression performance is affected by the overhead of metadata, this work re-generates the restore recipe using a chained tree structure, thus involving no extra storage overhead for compressed data, which substantially improves the compression ratios. We further derive a mathematical proof that lays the foundation for our method. The results demonstrate that zMesh can improve the smoothness of data by 67.9% and 71.3% for Z-ordering and Hilbert, respectively. Overall, zMesh improves the compression ratios by up to 16.5% and 133.7% for ZFP and SZ, respectively. Despite that zMesh involves additional compute overhead for tree and restore recipe construction, we show that the cost can be amortized as the number of quantities to be compressed increases.

The applications being developed within the U.S. Exascale Computing Project (ECP) to run on imminent Exascale computers will generate scientific results with unprecedented fidelity and record turn-around time. Many of these codes are based on particle-mesh methods and use advanced algorithms, especially dynamic load-balancing and mesh-refinement, to achieve high performance on Exascale machines. Yet, as such algorithms improve parallel application efficiency, they raise new challenges for I/O logic due to their irregular and dynamic data distributions. Thus, while the enormous data rates of Exascale simulations already challenge existing file system write strategies, the need for efficient read and processing of generated data introduces additional constraints on the data layout strategies that can be used when writing data to secondary storage. We review these I/O challenges and introduce two online data layout reorganization approaches for achieving good tradeoffs between read and write performance. We demonstrate the benefits of using these two approaches for the ECP particle-in-cell simulation WarpX, which serves as a motif for a large class of important Exascale applications. We show that by understanding application I/O patterns and carefully designing data layouts we can increase read performance by more than 80 percent.

Scientific simulations on high-performance computing systems produce vast amounts of data that need to be stored and analyzed efficiently. Lossy compression significantly reduces the data volume by trading accuracy for performance. Despite the recent success of lossy compression, such as ZFP and SZ, the compression performance is still far from being able to keep up with the exponential growth of data. This paper aims to further take advantage of application characteristics, an area that is often under-explored, to improve the compression ratios of adaptive mesh refinement (AMR) - a widely used numerical solver that allows for an improved resolution in limited regions. We propose a level reordering technique zMesh to reduce the storage footprint of AMR applications. In particular, we group the data points that are mapped to the same or adjacent geometric coordinates such that the dataset is smoother and more compressible. Unlike the prior work where the compression performance is affected by the overhead of metadata, this work re-generates restore recipe using a chained tree structure, thus involving no extra storage overhead for compressed data, which substantially improves the compression ratios. The results demonstrate that zMesh can improve the smoothness of data by 67.9% and 71.3% for Z-ordering and Hilbert, respectively. Overall, zMesh improves the compression ratios by up to 16.5% and 133.7% for ZFP and SZ, respectively. Despite that zMesh involves additional compute overhead for tree and restore recipe construction, we show that the cost can be amortized as the number of quantities to be compressed increases.

Extreme-scale simulations and high-resolution instruments have been generating an increasing amount of data, which poses significant challenges to not only data storage during the run, but also post-processing where data will be repeatedly retrieved and analyzed for a long period of time. The challenges in satisfying a wide range of post-hoc analysis needs while minimizing the I/O overhead caused by inappropriate and/or excessive data retrieval should never be left unmanaged. In this paper, we propose a data refactoring, compressing, and retrieval framework capable of 1) fine-grained data refactoring with regard to precision; 2) incrementally retrieving and recomposing the data in terms of various error bounds; and 3) adaptively retrieving data in multi-precision and multi-resolution with respect to different analysis. With the progressive data re-composition and the adaptable retrieval algorithms, our framework significantly reduces the amount of data retrieved when multiple incremental precision are requested and/or the downstream analysis time when coarse resolution is used. Experiments show that the amount of data retrieved under the same progressively requested error bound using our framework is 64% less than that using state-of-the-art single-error-bounded approaches. Parallel experiments with up to 1, 024 cores and ~ 600 GB data in total show that our approach yields 1.36× and 2.52× performance over existing approaches in writing to and reading from persistent storage systems, respectively.

With the increasing fidelity and resolution enabled by high-performance computing systems, simulation-based scientific discovery is able to model and understand microscopic physical phenomena at a level that was not possible in the past. A grand challenge that the HPC community is faced with is how to handle the large amounts of analysis data generated from simulations. In-memory computing, among others, is recognized to be a viable path forward and has experienced tremendous success in the past decade. Nevertheless, there has been a lack of a complete study and understanding of in-memory computing as a whole on HPC systems. This paper presents a comprehensive study, which goes well beyond the typical performance metrics. In particular, we assess the in-memory computing with regard to its usability, portability, robustness and internal design trade-offs, which are the key factors that of interest to domain scientists. We use two realistic scientific workflows, LAMMPS and Laplace, to conduct comprehensive studies on state-of-the-art in-memory computing libraries, including DataSpaces, DIMES, Flexpath and Decaf. We conduct cross-platform experiments at scale on two leading supercomputers, Titan at ORNL and Cori at NERSC, and summarize our key findings in this critical area.

As high-performance computing (HPC) is being scaled up to exascale to accommodate new modeling and simulation needs, I/O has continued to be a major bottleneck in the end-to-end scientific processes. Nevertheless, prior work in this area mostly aimed to maximize the average performance, and there has been a lack of study and solutions that can manage I/O performance variation on HPC systems. This work aims to take advantage of the storage characteristics and explore application level solutions that are interference-aware. In particular, we monitor the performance of data analytics and estimate the state of shared storage resources using discrete fourier transform (DFT). If heavy I/O interference is predicted to occur at a given timestep, data analytics can dynamically adapt to the environment by lowering the accuracy and performing partial or no augmentation from the shared storage, dictated by an augmentation-bandwidth plot. We evaluate three data analytics, XGC, GenASiS, and Jet, on Chameleon, and quantitatively demonstrate that both the average and variation of I/O performance can be vastly improved using our dynamic augmentation, with the mean and variance improved by as much as 67% and 96%, respectively, while maintaining acceptable outcome of data analysis.

This work presents a workflow for simulating and processing the full-scale low-frequency telescope data of the Square Kilometre Array (SKA) Phase 1. The SKA project will enter the construction phase soon, and once completed, it will be the world's largest radio telescope and one of the world's largest data generators. The authors used Summit to mimic an endto-end SKA workflow, simulating a dataset of a typical 6 hour observation and then processing that dataset with an imaging pipeline. This workflow was deployed and run on 4,560 compute nodes, and used 27,360 GPUs to generate 2.6 PB of data. This was the first time that radio astronomical data were processed at this scale. Results show that the workflow has the capability to process one of the key SKA science cases, an Epoch of Reionization observation. This analysis also helps reveal critical design factors for the next-generation radio telescopes and the required dedicated processing facilities.

This article studies the I/O write behaviors of the Titan supercomputer and its Lustre parallel file stores under production load. The results can inform the design, deployment, and configuration of file systems along with the design of I/O software in the application, operating system, and adaptive I/O libraries. We propose a statistical benchmarking methodology to measure write performance across I/O configurations, hardware settings, and system conditions. Moreover, we introduce two relative measures to quantify the write-performance behaviors of hardware components under production load. In addition to designing experiments and benchmarking on Titan, we verify the experimental results on one real application and one real application I/O kernel, XGC and HACC IO, respectively. These two are representative and widely used to address the typical I/O behaviors of applications. In summary, we find that Titan’s I/O system is variable across the machine at fine time scales. This variability has two major implications. First, stragglers lessen the benefit of coupled I/O parallelism (striping). Peak median output bandwidths are obtained with parallel writes to many independent files, with no striping or write sharing of files across clients (compute nodes). I/O parallelism is most effective when the application—or its I/O libraries—distributes the I/O load so that each target stores files for multiple clients and each client writes files on multiple targets in a balanced way with minimal contention. Second, our results suggest that the potential benefit of dynamic adaptation is limited. In particular, it is not fruitful to attempt to identify “good locations” in the machine or in the file system: component performance is driven by transient load conditions and past performance is not a useful predictor of future performance. For example, we do not observe diurnal load patterns that are predictable.

Scientific data collections grow ever larger, both in terms of the size of individual data items and of the number and complexity of items. To use and manage them, it is important to directly address issues of robust and actionable provenance. We identify three key drivers as our focus: managing the size and complexity of metadata, lack of a priori information to match usage intents between publishers and consumers of data, and support for campaigns over collections of data driven by multi-disciplinary, collaborating teams. We introduce the Hoarde abstraction as an attempt to formalize a way of looking at collections of data to make them more tractable for later use. Hoarde leverages middleware and systems infrastructures for scientific and technical data management. Through the lens of a select group of challenging data usage scenarios, we discuss some of the aspects of implementation, usage, and forward portability of this new view on data management.

For a production high-performance computing (HPC) system, where storage devices are shared between multiple applications and managed in a best effort manner, I/O contention is often a major problem. In this paper, we propose a balanced messaging-based re-routing in conjunction with throttling at the middleware level. This work tackles two key challenges that have not been fully resolved in the past: whether I/O variability can be reduced on a QoS-less HPC storage system, and how to design a runtime scheduling system that can scale up to a large amount of cores. The proposed scheme uses a two-level messaging system to re-route I/O requests to a less congested storage location so that write performance is improved, while limiting the impact on read by throttling re-routing. An analytical model is derived to guide the setup of optimal throttling factor. We thoroughly analyze the virtual messaging layer overhead and explore whether the in-transit buffering is effective in managing I/O variability. Contrary to the intuition, in-transit buffer cannot completely solve the problem. It can reduce the absolute variability but not the relative variability. The proposed scheme is verified against a synthetic benchmark as well as being used by production applications.

As a result of increasing data volume and velocity, Big Data science at exascale has shifted towards the in-situ paradigm, where large scale simulations run concurrently alongside data analytics. With in-situ, data generated from simulations can be processed while still in memory, thereby avoiding the slow storage bottleneck. However, running simulations and analytics together on shared resources will likely result in substantial contention if left unmanaged, as demonstrated in this work, leading to much reduced efficiency of simulations and analytics. Recently, virtualization technologies such as Linux containers have been widely applied to data centers and physical clusters to provide highly efficient and elastic resource provisioning for consolidated workloads including scientific simulations and data analytics. In this paper, we investigate to facilitate network traffic manipulation and reduce mutual interference on the network for in-situ applications in virtual clusters. In order to dynamically allocate the network bandwidth when it is needed, we adopt SARIMA-based techniques to analyze and predict MPI traffic issued from simulations. Although this can be an effective technique, the naïve usage of network virtualization can lead to performance degradation for bursty asynchronous transmissions within an MPI job. We analyze and resolve this performance degradation in virtual clusters.

With the growing computational complexity of science and the complexity of new and emerging hardware, it is time to re-evaluate the traditional monolithic design of computational codes. One new paradigm is constructing larger scientific computational experiments from the coupling of multiple individual scientific applications, each targeting their own physics, characteristic lengths, and/or scales. We present a framework constructed by leveraging capabilities such as in-memory communications, workflow scheduling on HPC resources, and continuous performance monitoring. This code coupling capability is demonstrated by a fusion science scenario, where differences between the plasma at the edges and at the core of a device have different physical descriptions. This infrastructure not only enables the coupling of the physics components, but it also connects in situ or online analysis, compression, and visualization that accelerate the time between a run and the analysis of the science content. Results from runs on Titan and Cori are presented as a demonstration.

One of the core issues across computer and computational science today is adapting to, managing, and learning from the influx of "Big Data". In the commercial space, this problem has led to a huge investment in new technologies and capabilities that are well adapted to dealing with the sorts of human-generated logs, videos, texts, and other large-data artifacts that are processed and resulted in an explosion of useful platforms and languages (Hadoop, Spark, Pandas, etc.). However, translating this work from the enterprise space to the computational science and HPC community has proven somewhat difficult, in part because of some of the fundamental differences in type and scale of data and timescales surrounding its generation and use. We describe a forward-looking research and development plan which centers around the concept of making Input/Output (I/O) intelligent for users in the scientific community, whether they are accessing scalable storage or performing in situ workflow tasks. Much of our work is based on our experience with the Adaptable I/O System (ADIOS 1.X), and our next generation version of the software ADIOS 2.X [1].

Scientific simulations generate large amounts of floating-point data, which are often not very compressible using the traditional reduction schemes, such as deduplication or lossless compression. The emergence of lossy floating-point compression holds promise to satisfy the data reduction demand from HPC applications; however, lossy compression has not been widely adopted in science production. We believe a fundamental reason is that there is a lack of understanding of the benefits, pitfalls, and performance of lossy compression on scientific data. In this paper, we conduct a comprehensive study on state-of-the-art lossy compression, including ZFP, SZ, and ISABELA, using real and representative HPC datasets. Our evaluation reveals the complex interplay between compressor design, data features and compression performance. The impact of reduced accuracy on data analytics is also examined through a case study of fusion blob detection, offering domain scientists with the insights of what to expect from fidelity loss. Furthermore, the trial and error approach to understanding compression performance involves substantial compute and storage overhead. To this end, we propose a sampling based estimation method that extrapolates the reduction ratio from data samples, to guide domain scientists to make more informed data reduction decisions.

Scientific discoveries are increasingly relying on analysis of massive amounts of data. The ability to directly access the most relevant data records through query, without shifting through all of them becomes essential. However, scientific datasets are commonly stored on parallel file systems and I/O systems that are optimized for reading/writing large chunks of data, and many scientific datasets have spatial-temporal data similarity, such that the records with similar values often locate in a close proximity of each other. Therefore, our previous work started to investigate the benefit of using block range index technique for scientific datasets, which only records the value range of all the records in a data block. In this paper, we extend our work in several aspects. First, we implement and integrate our blockindex technique with the ADIOS I/O system. Second, we show our proposed method can be significantly better than the existing minmax and bitmaps indexing methods supported in ADIOS, and can also have comparable performance in the worst case. Third, we propose several techniques that can take advantage of the block index information to greatly reduce data retrieval time from query results. Fourth, we evaluate our approach using several real scientific datasets, and analyze the spatial-temporal data similarity characteristics in them. Through our study, we believe block index can be an effective indexing technique for scientific datasets with little implementation and operating overhead. It's size is small enough for building the indexes on-the-fly, and yet its query information is sufficient for efficient data access.

Task mapping is an important problem in parallel and distributed computing. The goal in task mapping is to find an optimal layout of the processes of an application (or a task) onto a given network topology. We target this problem in the context of staging applications. A staging application consists of two or more parallel applications (also referred to as staging tasks) which run concurrently and exchange data over the course of computation. Task mapping becomes a more challenging problem in staging applications, because not only data is exchanged between the staging tasks, but also the processes of a staging task may exchange data with each other. We propose a novel method, called Task Graph Embedding (TGE), that harnesses the observable graph structures of parallel applications and network topologies. TGE employs a machine learning based algorithm to find the best representation of a graph, called an embedding, onto a space in which the task-to-processor mapping problem can be solved. We evaluate and demonstrate the effectiveness of TGE experimentally with the communication patterns extracted from runs of XGC, a large-scale fusion simulation code, on Titan.

As the memory and storage hierarchy get deeper and more complex, it is important to have new benchmarks and evaluation tools that allow us to explore the emerging middleware solutions to use this hierarchy. Skel is a tool aimed at automating and refining this process of studying HPC I/O performance. It works by generating application I/O kernel/benchmarks as determined by a domain-specific model. This paper provides some techniques for extending Skel to address new situations and to answer new research questions. For example, we document use cases as diverse as using Skel to troubleshoot I/O performance issues for remote users, refining an I/O system model, and facilitating the development and testing of a mechanism for runtime monitoring and performance analytics. We also discuss data oriented extensions to Skel to support the study of compression techniques for Exascale scientific data management.

Scientific simulations on high performance computing (HPC) platforms generate large quantities of data. To bridge the widening gap between compute and I/O, and enable data to be more efficiently stored and analyzed, simulation outputs need to be refactored, reduced, and appropriately mapped to storage tiers. However, a systematic solution to support these steps has been lacking on the current HPC software ecosystem. To that end, this paper develops Canopus, a progressive JPEGlike data management scheme for storing and analyzing big scientific data. It co-designs the data decimation, compression and data storage, taking the hardware characteristics of each storage tier into considerations. With reasonably low overhead, our approach refactors simulation data into a much smaller, reduced-accuracy base dataset, and a series of deltas that is used to augment the accuracy if needed. The base dataset and deltas are compressed and written to multiple storage tiers. Data saved on different tiers can then be selectively retrieved to restore the level of accuracy that satisfies data analytics. Thus, Canopus provides a paradigm shift towards elastic data analytics and enables end users to make trade-offs between analysis speed and accuracy on-the-fly. We evaluate the impact of Canopus on unstructured triangular meshes, a pervasive data model used by scientific modeling and simulations. In particular, we demonstrate the progressive data exploration of Canopus using the “blob detection” use case on the fusion simulation data.

As we continue toward exascale, scientific data volume is continuing to scale and becoming more burdensome to manage. In this paper, we lay out opportunities to enhance state of the art data management techniques. We emphasize well-principled data compression, and using it to achieve progressive refinement. This can both accelerate I/O and afford the user increased flexibility when she interacts with the data. The formulation naturally maps onto enabling partitioning of the progressively improving-quality representations of a data quantity into different media-type destinations, to keep the highest priority information as close as possible to the computation, and take advantage of deepening memory/storage hierarchies in ways not previously possible. Careful monitoring is requisite to our vision, not only to verify that compression has not eliminated salient features in the data, but also to better understand the performance of massively parallel scientific applications. Increased mathematical rigor would be ideal,to help bring compression on a better-understood theoretical footing, closer to the relevant scientific theory, more aware of constraints imposed by the science, and more tightly error-controlled. Throughout, we highlight pathfinding research we have begun exploring related these topics, and comment toward future work that will be needed.

Die-stacked DRAM (a.k.a., on-chip DRAM) provides much higher bandwidth and lower latency than off-chip DRAM. It is a promising technology to break the "memory wall". Die-stacked DRAM can be used either as a cache (i.e., DRAM cache) or as a part of memory (PoM). A DRAM cache design would suffer from more page faults than a PoM design as the DRAM cache cannot contribute towards capacity of main memory. At the same time, obtaining high performance requires PoM systems to swap requested data to the die-stacked DRAM. Existing PoM designs fall into two categories – line-based and page-based. The former ensures low off-chip bandwidth utilization but suffers from a low hit ratio of on-chip memory due to limited temporal locality. In contrast, page-based designs achieve a high hit ratio of on-chip memory albeit at the cost of moving large amounts of data between on-chip and off-chip memories, leading to increased off-chip bandwidth utilization and significant system performance degradation.To achieve a similar high hit ratio of on-chip memory as page-based designs, and eliminate excessive off-chip traffic involved, we propose SELF, a high performance and bandwidth efficient approach. The key idea is to SElectively swap Lines in a requested page that are likely to be accessed according to page Footprint, instead of blindly swapping an entire page. In doing so, SELF allows incoming requests to be serviced from the on-chip memory as much as possible, while avoiding swapping unused lines to reduce memory bandwidth consumption. We evaluate a memory system which consists of 4GB on-chip DRAM and 12GB off-chip DRAM. Compared to a baseline system that has the same total capacity of 16GB off-chip DRAM, SELF improves the performance in terms of instructions per cycle by 26.9%, and reduces the energy consumption per memory access by 47.9% on average. In contrast, state-of-the-art line-based and page-based PoM designs can only improve the performance by 9.5% and 9.9%, respectively, against the same baseline system.

As applications target extreme scales, data staging and in-situ/in-transit data processing have been proposed to address the data challenges and improve scientific discovery. However, further research is necessary in order to understand how growing data sizes from data intensive simulations coupled with the limited DRAM capacity in High End Computing systems will impact the effectiveness of this approach. In this paper, we explore how we can use deep memory levels for data staging, and develop a multi-tiered data staging method that spans bothDRAM and solid state disks (SSD). This approach allows us to support both code coupling and data management for data intensive simulation workflows. We also show how an adaptive application-aware data placement mechanism can dynamically manage and optimize data placement across the DRAM ands storage levels in this multi-tiered data staging method. We present an experimental evaluation of our approach using wolf resources: an Infiniband cluster (Sith) and a Cray XK7system (Titan), and using combustion (S3D) and fusion (XGC1) simulations.

Parallel application benchmarks are indispensable for evaluating/optimizing HPC software and hardware. However, it is very challenging and costly to obtain high-fidelity benchmarks reflecting the scale and complexity of state-of-the-art parallel applications. Hand-extracted synthetic benchmarks are time- and labor-intensive to create. Real applications themselves, while offering most accurate performance evaluation, are expensive to compile, port, recon- figure, and often plainly inaccessible due to security or ownership concerns. This work contributes APPRIME, a novel tool for trace-based automatic parallel benchmark generation. Taking as input standard communication-I/O traces of an application’s execution, it couples accurate automatic phase identification with statistical regeneration of event parameters to create compact, portable, and to some degree reconfigurable parallel application benchmarks. Experiments with four NAS Parallel Benchmarks (NPB) and three real scientific simulation codes confirm the fidelity of APPRIME benchmarks. They retain the original applications’ performance characteristics, in particular the relative performance across platforms.

Parallel application benchmarks are indispensable for evaluating/optimizing HPC software and hardware. However, it is very challenging and costly to obtain high-fidelity benchmarks reflecting the scale and complexity of state-of-the-art parallel applications. Hand-extracted synthetic benchmarks are time- and labor-intensive to create. Real applications themselves, while offering most accurate performance evaluation, are expensive to compile, port, reconfigure, and often plainly inaccessible due to security or ownership concerns. This work contributes APPrime, a novel tool for trace-based automatic parallel benchmark generation. Taking as input standard communication-I/O traces of an application's execution, it couples accurate automatic phase identification with statistical regeneration of event parameters to create compact, portable, and to some degree reconfigurable parallel application benchmarks. Experiments with four NAS Parallel Benchmarks (NPB) and three real scientific simulation codes confirm the fidelity of APPrime benchmarks. They retain the original applications' performance characteristics, in particular their relative performance across platforms. Also, the result benchmarks, already released online, are much more compact and easy-to-port compared to the original applications.

Summary Managing the large volumes of data produced by emerging scientific and engineering simulations running on leadership‐class resources has become a critical challenge. The data have to be extracted off the computing nodes and transported to consumer nodes so that it can be processed, analyzed, visualized, archived, and so on. Several recent research efforts have addressed data‐related challenges at different levels. One attractive approach is to offload expensive input/output operations to a smaller set of dedicated computing nodes known as a staging area. However, even using this approach, the data still have to be moved from the staging area to consumer nodes for processing, which continues to be a bottleneck. In this paper, we investigate an alternate approach, namely moving the data‐processing code to the staging area instead of moving the data to the data‐processing code. Specifically, we describe the ActiveSpaces framework, which provides (1) programming support for defining the data‐processing routines to be downloaded to the staging area and (2) runtime mechanisms for transporting codes associated with these routines to the staging area, executing the routines on the nodes that are part of the staging area, and returning the results. We also present an experimental performance evaluation of ActiveSpaces using applications running on the Cray XT5 at Oak Ridge National Laboratory. Finally, we use a coupled fusion application workflow to explore the trade‐offs between transporting data and transporting the code required for data processing during coupling, and we characterize sweet spots for each option. Copyright © 2014 John Wiley & Sons, Ltd.

Though an abundance of novel "data transformation" technologies have been developed (such as compression, level-of-detail, layout optimization, and indexing), there remains a notable gap in the adoption of such services by scientific applications. In response, we develop an in situ data transformation framework in the ADIOS I/O middleware with a "plug in" interface, thus greatly simplifying both the deployment and use of data transform services in scientific applications. Our approach ensures user-transparency, runtime-configurability, compatibility with existing I/O optimizations, and the potential for exploiting read-optimizing transforms (such as level-of-detail) to achieve I/O reduction. We demonstrate use of our framework with the QLG simulation at up to 8,192 cores on the leadership-class Titan supercomputer, showing negligible overhead. We also explore the read performance implications of data transforms with respect to parameters such as chunk size, access pattern, and the "opacity" of different transform methods including compression and level-of-detail.

As high-end systems move toward exascale sizes, a new model of scientific inquiry being developed is one in which online data analytics run concurrently with the high end simulations producing data outputs. Goals are to gain rapid insights into the ongoing scientific processes, assess their scientific validity, and/or initiate corrective or supplementary actions by launching additional computations when needed. The Flex path system presented in this paper addresses the fundamental problem of how to structure and efficiently implement the communications between high end simulations and concurrently running online data analytics, the latter comprised of componentized dynamic services and service pipelines. Using a type-based publish/subscribe approach, Flexpath encourages diversity by permitting analytics services to differ in their computational and scaling characteristics and even in their internal execution models. Flex path uses direct and MxN connections between interacting services to reduce data movements, to allow for runtime connectivity changes to accommodate component arrivals/departures, and to support the multiple underlying communication protocols used for analytics workflows in which simulation outputs are processed by analytics services residing on the same nodes where they are generated, on the same machine, and/or on attached or remote analytics engines. This paper describes the design and implementation of Flex path, and evaluates it with two widely used scientific applications and their associated data analytics methods.

Next generation in-situ/in-transit data processing has been proposed for addressing data challenges at extreme scales. However, further research is necessary in order to understand how growing data sizes from data intensive simulations coupled with limited DRAM capacity in High End Computing clusters will impact the effectiveness of this approach. In this work, we propose using deep memory levels for data staging, utilizing a multi-tiered data staging method with both DRAM and solid state disk (SSD). This approach allows us to support both code coupling and data management for data intensive simulations in cluster environment. We also show how an application-aware data placement mechanism can dynamically manage and optimize data placement across DRAM and SSD storage levels in staging method. We present experimental results on Sith - an Infiniband cluster at Oak Ridge, and evaluate its performance using combustion (S3D) and fusion (XGC) simulations.

On petascale systems, the selection of optimal values for I/O parameters without taking into account the I/O size and pattern can cause the I/O time to dominate the simulation time, compromising the application's scalability. In this paper, we adopt and adapt an engineering method called surrogate-based modeling to efficiently search for the optimal I/O parameter values and accurately predict the associated I/O times at the extreme scale. Our approach allows us to address both the search and prediction in a short time, even when the application's I/O is large and exhibits irregular patterns.

SUMMARY Applications running on leadership platforms are more and more bottlenecked by storage input/output (I/O). In an effort to combat the increasing disparity between I/O throughput and compute capability, we created Adaptable IO System (ADIOS) in 2005. Focusing on putting users first with a service oriented architecture, we combined cutting edge research into new I/O techniques with a design effort to create near optimal I/O methods. As a result, ADIOS provides the highest level of synchronous I/O performance for a number of mission critical applications at various Department of Energy Leadership Computing Facilities. Meanwhile ADIOS is leading the push for next generation techniques including staging and data processing pipelines. In this paper, we describe the startling observations we have made in the last half decade of I/O research and development, and elaborate the lessons we have learned along this journey. We also detail some of the challenges that remain as we look toward the coming Exascale era. Copyright © 2013 John Wiley & Sons, Ltd.

The size and scope of cutting-edge scientific simulations are growing much faster than the I/O and storage capabilities of their run-time environments. The growing gap is exacerbated by exploratory, data-intensive analytics, such as querying simulation data with multivariate, spatio-temporal constraints, which induces heterogeneous access patterns that stress the performance of the underlying storage system. Previous work addresses data layout and indexing techniques to improve query performance for a single access pattern, which is not sufficient for complex analytics jobs. We present PARLO a parallel run-time layout optimization framework, to achieve multi-level data layout optimization for scientific applications at run-time before data is written to storage. The layout schemes optimize for heterogeneous access patterns with user-specified priorities. PARLO is integrated with ADIOS, a high-performance parallel I/O middleware for large-scale HPC applications, to achieve user-transparent, light-weight layout optimization for scientific datasets. It offers simple XML-based configuration for users to achieve flexible layout optimization without the need to modify or recompile application codes. Experiments show that PARLO improves performance by 2 to 26 times for queries with heterogeneous access patterns compared to state-of-the-art scientific database management systems. Compared to traditional post-processing approaches, its underlying run-time layout optimization achieves a 56% savings in processing time and a reduction in storage overhead of up to 50%. PARLO also exhibits a low run-time resource requirement, while also limiting the performance impact on running applications to a reasonable level.

Scientific communities have benefitted from a significant increase of available computing and storage resources in the last few decades. For science projects that have access to leadership scale computing resources, the capacity to produce data has been growing exponentially. Teams working on such projects must now include, in addition to the traditional application scientists, experts in various disciplines including applied mathematicians for development of algorithms, visualization specialists for large data, and I/O specialists. Sharing of knowledge and data is becoming a requirement for scientific discovery, providing useful mechanisms to facilitate this sharing is a key challenge for e-Science. Our hypothesis is that in order to decrease the time to solution for application scientists we need to lower the barrier of entry into related computing fields. We aim at improving users' experience when interacting with a vast software ecosystem and/or huge amount of data, while maintaining focus on their primary research field. In this context we present our approach to bridge the gap between the application scientists and the visualization experts through a visualization schema as a first step and proof of concept for a new way to look at interdisciplinary collaboration among scientists dealing with big data. The key to our approach is recognizing that our users are scientists who mostly work as islands. They tend to work in very specialized environment but occasionally have to collaborate with other researchers in order to take full advantage of computing innovations and get insight from big data. We present an example of identifying the connecting elements between one of such relationships and offer a liaison schema to facilitate their collaboration.

Analysis of large pathology image datasets offers significant opportunities for the investigation of disease morphology, but the resource requirements of analysis pipelines limit the scale of such studies. Motivated by a brain cancer study, we propose and evaluate a parallel image analysis application pipeline for high throughput computation of large datasets of high resolution pathology tissue images on distributed CPU-GPU platforms. To achieve efficient execution on these hybrid systems, we have built runtime support that allows us to express the cancer image analysis application as a hierarchical data processing pipeline. The application is implemented as a coarse-grain pipeline of stages, where each stage may be further partitioned into another pipeline of fine-grain operations. The fine-grain operations are efficiently managed and scheduled for computation on CPUs and GPUs using performance aware scheduling techniques along with several optimizations, including architecture aware process placement, data locality conscious task assignment, data prefetching, and asynchronous data copy. These optimizations are employed to maximize the utilization of the aggregate computing power of CPUs and GPUs and minimize data copy overheads. Our experimental evaluation shows that the cooperative use of CPUs and GPUs achieves significant improvements on top of GPU-only versions (up to 1.6×) and that the execution of the application as a set of fine-grain operations provides more opportunities for runtime optimizations and attains better performance than coarser-grain, monolithic implementations used in other works. An implementation of the cancer image analysis pipeline using the runtime support was able to process an image dataset consisting of 36,848 4Kx4K-pixel image tiles (about 1.8TB uncompressed) in less than 4 minutes (150 tiles/second) on 100 nodes of a state-of-the-art hybrid cluster system.

Increasingly severe I/O bottlenecks on High-End Computing machines are prompting scientists to process simulation output data online while simulations are running and before storing data on disk. There are several options to place data analytics along the I/O path: on compute nodes, on separate nodes dedicated to analytics, or after data is stored on persistent storage. Since different placements have different impact on performance and cost, there is a consequent need for flexibility in the location of data analytics. The FlexIO middleware described in this paper makes it easy for scientists to obtain such flexibility, by offering simple abstractions and diverse data movement methods to couple simulation with analytics. Various placement policies can be built on top of FlexIO to exploit the trade-offs in performing analytics at different levels of the I/O hierarchy. Experimental results demonstrate that FlexIO can support a variety of simulation and analytics workloads at large scale through flexible placement options, efficient data movement, and dynamic deployment of data manipulation functionalities.

In the era of petascale computing, more scientific applications are being deployed on leadership scale computing platforms to enhance the scientific productivity. Many I/O techniques have been designed to address the growing I/O bottleneck on large-scale systems by handling massive scientific data in a holistic manner. While such techniques have been leveraged in a wide range of applications, they have not been shown as adequate for many mission critical applications, particularly in data postprocessing stage. One of the examples is that some scientific applications generate datasets composed of a vast amount of small data elements that are organized along many spatial and temporal dimensions but require sophisticated data analytics on one or more dimensions. Including such dimensional knowledge into data organization can be beneficial to the efficiency of data post-processing, which is often missing from exiting I/O techniques. In this study, we propose a novel I/O scheme named STAR (Spatial and Temporal AggRegation) to enable high performance data queries for scientific analytics. STAR is able to dive into the massive data, identify the spatial and temporal relationships among data variables, and accordingly organize them into an optimized multi-dimensional data structure before storing to the storage. This technique not only facilitates the common access patterns of data analytics, but also further reduces the application turnaround time. In particular, STAR is able to enable efficient data queries along the time dimension, a practice common in scientific analytics but not yet supported by existing I/O techniques. In our case study with a critical climate modeling application GEOS-5, the experimental results on Jaguar supercomputer demonstrate an improvement up to 73 times for the read performance compared to the original I/O method.

Fast growing large-scale systems enable scientific applications to run at a much larger scale and accordingly produce gigantic volumes of simulation output. Such data imposes a grand challenge to post-processing tasks such as visualization and data analysis, because these tasks are often performed at a host machine that is remotely located and equipped with much less memory and storage resources. During the simulation runs, it is also desirable for scientists to be able to interactively monitor and steer the progress of simulation. This requires scientific data to be represented in an efficient form for initial exploration and computation steering. In this paper, we propose DynaM a software framework that can represent scientific data in a multiresolution form, and dynamically organize data blocks into an optimized layout for efficient scientific analysis. DynaM supports a convolution-based multiresolution data representation for abstracting scientific data for visualization at a wide spectrum of resolution. To support the efficient generation and retrieval of different data granularities from such representation, a dynamic data organization in DynaM is enabled to cater distinct peculiarities of different size data blocks for efficient and balanced I/O performance. Our experimental results demonstrate that DynaM can efficiently represent large scientific dataset and speed up the visualization of multidimensional scientific data. An up to 29 times speedup is achieved on Jaguar supercomputer at Oak Ridge National Laboratory.

As system scales and application complexity grow, managing and processing simulation data has become a significant challenge. While recent approaches based on data staging and in-situ/in-transit data processing are promising, dynamic data volumes and distributions, such as those occurring in AMR-based simulations, make the efficient use of these techniques challenging. In this paper we propose cross-layer adaptations that address these challenges and respond at runtime to dynamic data management requirements. Specifically we explore (1) adaptations of the spatial resolution at which the data is processed, (2) dynamic placement and scheduling of data processing kernels, and (3) dynamic allocation of in-transit resources. We also exploit coordinated approaches that dynamically combine these adaptations at the different layers. We evaluate the performance of our adaptive cross-layer management approach on the Intrepid IBM-BlueGene/P and Titan Cray-XK7 systems using Chombo-based AMR applications, and demonstrate its effectiveness in improving overall time-to-solution and increasing resource efficiency.

Predictive pre-fetcher, which predicts future data access events and loads the data before users requests, has been widely studied, especially in file systems or web contents servers, to reduce data load latency. Especially in scientific data visualization, pre-fetching can reduce the IO waiting time. In order to increase the accuracy, we apply a data mining technique to extract hidden information. More specifically, we apply a data mining technique for discovering the hidden contexts in data access patterns and make prediction based on the inferred context to boost the accuracy. In particular, we performed Probabilistic Latent Semantic Analysis (PLSA), a mixture model based algorithm popular in the text mining area, to mine hidden contexts from the collected user access patterns and, then, we run a predictor within the discovered context. We further improve PLSA by applying the Deterministic Annealing (DA) method to overcome the local optimum problem. In this paper we demonstrate how we can apply PLSA and DA optimization to mine hidden contexts from users data access patterns and improve predictive pre-fetcher performance.

Emerging scientific and engineering simulations running at scale on leadership-class High End Computing (HEC) environments are producing large volumes of data, which has to be transported and analyzed before any insights can result from these simulations. The complexity and cost (in terms of time and energy) associated with managing and analyzing this data have become significant challenges, and are limiting the impact of these simulations. Recently, data-staging approaches along with in-situ and in-transit analytics have been proposed to address these challenges by offloading I/O and/or moving data processing closer to the data. However, scientists continue to be overwhelmed by the large data volumes and data rates. In this paper we address this latter challenge. Specifically, we propose a highly scalable and low-overhead associative messaging framework that runs on the data staging resources within the HEC platform, and builds on the staging-based online in-situ/in-transit analytics to provide publish/subscribe/notification-type messaging patterns to the scientist. Rather than having to ingest and inspect the data volumes, this messaging system allows scientists to (1) dynamically subscribe to data events of interest, e.g., simple data values or a complex function or simple reduction (max()/min()/avg()) of the data values in a certain region of the application domain is greater/less than a threshold value, or certain spatial/temporal data features or data patterns are detected; (2) define customized in-situ/in-transit actions that are triggered based on the events, such as data visualization or transformation; and (3) get notified when these events occur. The key contribution of this paper is a design and implementation that can support such a messaging abstraction at scale on high-end computing (HEC) systems with minimal overheads. We have implemented and deployed the messaging system on the Jaguar Cray XK6 machines at Oak Ridge National Laboratory and the Lonestar system at the Texas Advanced Computing Center (TACC), and we present the experimental performance evaluation using these HEC platforms in the paper.

Large-scale scientific applications on High End Computing systems produce a large volume of highly complex datasets. Such data imposes a grand challenge to conventional storage systems for the need of efficient I/O solutions during both the simulation runtime and data post-processing phases. With the mounting needs of scientific discovery, the read performance of large-scale simulations has becomes a critical issue for the HPC community. In this study, we propose a system-aware optimized data organization strategy that can organize data blocks of multidimensional scientific data efficiently based on simulation output and the underlying storage systems, thereby enabling efficient scientific analytics. Our experimental results demonstrate a performance speedup up to 72 times for the combustion simulation S3D, compared to the logically contiguous data layout.

Emerging scientific application workflows are composed of heterogeneous coupled component applications that simulate different aspects of the physical phenomena being modeled, and that interact and exchange significant volumes of data at runtime. With the increasing performance gap between on-chip data sharing and off-chip data transfers in current systems based on multicore processors, moving large volumes of data using communication network fabric can significantly impact performance. As a result, minimizing the amount of inter-application data exchanges that are across compute nodes and use the network is critical to achieving overall application performance and system efficiency. In this paper, we investigate the in-situ execution of the coupled components of a scientific application workflow so as to maximize on-chip exchange of data. Specifically, we present a distributed data sharing and task execution framework that (1) employs data-centric task placement to map computations from the coupled applications onto processor cores so that a large portion of the data exchanges can be performed using the intra-node shared memory, (2) provides a shared space programming abstraction that supplements existing parallel programming models (e.g., message passing) with specialized one-sided asynchronous data access operators and can be used to express coordination and data exchanges between the coupled components. We also present the implementation of the framework and its experimental evaluation on the Jaguar Cray XT5 at Oak Ridge National Laboratory.

Current I/O techniques have pushed the write performance close to the system peak, but they usually overlook the read side of problem. With the mounting needs of scientific discovery, it is important to provide good read performance for many common access patterns. Such demand requires an organization scheme that can effectively utilize the underlying storage system. However, the mismatch between conventional data layout on disk and common scientific access patterns leads to significant performance degradation when a subset of data is accessed. To this end, we design a system-aware Optimized Chunking model, which aims to find an optimized organization that can strike for a good balance between data transfer efficiency and processing overhead. To enable such model for scientific applications, we propose SMART-IO, a two-level data organization framework that can organize the blocks of multidimensional data efficiently. This scheme can adapt data layouts based on data characteristics and underlying storage systems, and enable efficient scientific analytics. Our experimental results demonstrate that SMART-IO can significantly improve the read performance for challenging access patterns, and speed up data analytics. For a mission critical combustion simulation code S3D, Smart-IO achieves up to 72 times speedup for planar reads of a 3-D variable compared to the logically contiguous data layout.

Supercomputer I/O loads are often dominated by writes. HPC (High Performance Computing) file systems are designed to absorb these bursty outputs at high bandwidth through massive parallelism. However, the delivered write bandwidth often falls well below the peak. This paper characterizes the data absorption behavior of a center-wide shared Lustre parallel file system on the Jaguar supercomputer. We use a statistical methodology to address the challenges of accurately measuring a shared machine under production load and to obtain the distribution of bandwidth across samples of compute nodes, storage targets, and time intervals. We observe and quantify limitations from competing traffic, contention on storage servers and I/O routers, concurrency limitations in the client compute node operating systems, and the impact of variance (stragglers) on coupled output such as striping. We then examine the implications of our results for application performance and the design of I/O middleware systems on shared supercomputers.

Large scale scientific applications are often bottlenecked due to the writing of checkpoint-restart data. Much work has been focused on improving their write performance. With the mounting needs of scientific discovery from these datasets, it is also important to provide good read performance for many common access patterns, which requires effective data organization. To address this issue, we introduce Elastic Data Organization (EDO), which can transparently enable different data organization strategies for scientific applications. Through its flexible data ordering algorithms, EDO harmonizes different access patterns with the underlying file system. Two levels of data ordering are introduced in EDO. One works at the level of data groups (a.k.a process groups). It uses Hilbert Space Filling Curves (SFC) to balance the distribution of data groups across storage targets. Another governs the ordering of data elements within a data group. It divides a data group into sub chunks and strikes a good balance between the size of sub chunks and the number of seek operations. Our experimental results demonstrate that EDO is able to achieve balanced data distribution across all dimensions and improve the read performance of multidimensional datasets in scientific applications.

Scientific applications are striving to accurately simulate multiple interacting physical processes that comprise complex phenomena being modeled. Efficient and scalable parallel implementations of these coupled simulations present challenging interaction and coordination requirements, especially when the coupled physical processes are computationally heterogeneous and progress at different speeds. In this paper, we present the design, implementation and evaluation of a memory-to-memory coupling framework for coupled scientific simulations on high-performance parallel computing platforms. The framework is driven by the coupling requirements of the Center for Plasma Edge Simulation, and it provides simple coupling abstractions as well as efficient asynchronous (RDMA-based) memory-to-memory data transport mechanisms that complement existing parallel programming systems and data sharing frameworks. The framework enables flexible coupling behaviors that are asynchronous in time and space, and it supports dynamic coupling between heterogeneous simulation processes without enforcing any synchronization constraints. We evaluate the performance and scalability of the coupling framework using a specific coupling scenario, on the Jaguar Cray XT5 system at Oak Ridge National Laboratory.

Peta-scale scientific applications running on High End Computing (HEC) platforms can generate large volumes of data. For high performance storage and in order to be useful to science end users, such data must be organized in its layout, indexed, sorted, and otherwise manipulated for subsequent data presentation, visualization, and detailed analysis. In addition, scientists desire to gain insights into selected data characteristics `hidden' or `latent' in these massive datasets while data is being produced by simulations. PreDatA, short for Preparatory Data Analytics, is an approach to preparing and characterizing data while it is being produced by the large scale simulations running on peta-scale machines. By dedicating additional compute nodes on the machine as `staging' nodes and by staging simulations' output data through these nodes, PreDatA can exploit their computational power to perform select data manipulations with lower latency than attainable by first moving data into file systems and storage. Such intransit manipulations are supported by the PreDatA middleware through asynchronous data movement to reduce write latency, application-specific operations on streaming data that are able to discover latent data characteristics, and appropriate data reorganization and metadata annotation to speed up subsequent data access. PreDatA enhances the scalability and flexibility of the current I/O stack on HEC platforms and is useful for data pre-processing, runtime data analysis and inspection, as well as for data exchange between concurrently running simulations.

The purpose of the Fusion Simulation Project is to develop a predictive capability for integrated modeling of magnetically confined burning plasmas. In support of this mission, the Center for Plasma Edge Simulation has developed an End-to-end Framework for Fusion Integrated Simulation (EFFIS) that combines critical computer science technologies in an effective manner to support leadership class computing and the coupling of complex plasma physics models. We describe here the main components of EFFIS and how they are being utilized to address our goal of integrated predictive plasma edge simulation.

Simulations that require massive amounts of computing power and generate tens of terabytes of data are now part of the daily lives of scientists. Analyzing and visualizing the results of these simulations as they are computed can lead not only to early insights but also to useful knowledge that can be provided as feedback to the simulation, avoiding unnecessary use of computing power. Our work is aimed at making advanced visualization tools available to scientists in a user-friendly, Web-based environment where they can be accessed anytime from anywhere. In the context of turbulent combustion for example, visualization is used to understand the coupling between turbulence and the turbulent mixing of scalars. Although isosurface generation is a useful technique in this scenario, computing and rendering isosurfaces one at a time is expensive and not particularly well-suited for such a Web-based framework. In this paper we propose the use of a summary structure, called contour tree, that captures the topological structure of a scalar field and guides the user in identifying useful isosurfaces. We have also designed an interface which has been integrated with a Web-based simulation monitoring system, that allows users to interact with and explore multiple isosurfaces.

Abstract Scientific workflows often benefit from or even require advanced modeling constructs, e.g. nesting of subworkflows, cycles for executing loops, data‐dependent routing, and pipelined execution. In such settings, an often overlooked aspect of provenance takes center stage: a suitable model of provenance (MoP) for scientific workflows should be based upon the underlying model of computation (MoC) used for executing the workflows. We can derive an adequate MoP from a MoC (such as Kahn's process networks) by taking into account the assumptions that a MoC entails, and by recording the observables which it affords. In this way, a MoP captures or at least better approximates ‘real’ data dependencies for workflows with advanced modeling constructs. As a specific instance, we elaborate on the Read–Write–ReSet model, a simple and flexible MoP suitable for a number of different MoCs. Copyright © 2007 John Wiley & Sons, Ltd.

Abstract The first Provenance Challenge was set up in order to provide a forum for the community to understand the capabilities of different provenance systems and the expressiveness of their provenance representations. To this end, a functional magnetic resonance imaging workflow was defined, which participants had to either simulate or run in order to produce some provenance representation, from which a set of identified queries had to be implemented and executed. Sixteen teams responded to the challenge, and submitted their inputs. In this paper, we present the challenge workflow and queries, and summarize the participants' contributions. Copyright © 2007 John Wiley & Sons, Ltd.

So far BOINC based desktop grid systems have been applied at the global computing level. This paper describes an extended version of BOINC called SZTAKI desktop grid (SZDG) that aims at using desktop grids (DGs) at local (enterprise/institution) level. The novelty of SZDG is that it enables the hierarchical organisation of local DGs, i.e., clients of a DG can be DGs at a lower level that can take work units from their higher level DG server. More than that, even clusters can be connected at the client level and hence work units can contain complete MPI programs to be run on the client clusters. In order to easily create master/worker type DG applications a new API, called as the DC-API has been developed. SZDG and DC-API has been successfully applied both at the global and local level, both in academic institutions and in companies to solve problems requiring large computing power.

We view scientific workflows as the domain scientist's way to harness cyberinfrastructure for e-Science. Domain scientists are often interested in "end-to-end" frameworks which include data acquisition, transformation, analysis, visualization, and other steps. While there is no lack of technologies and standards to choose from, a simple, unified framework combining data modeling and processoriented modeling and design of scientific workflows has yet to emerge. Towards this end, we introduce a number of concepts such as models of computation and provenance, actor-oriented modeling, adapters, hybrid types, and higher-order components, and then outline a particular composition of some of these concepts, yielding a promising new synthesis for describing scientific workflows, i.e., Collection-Oriented Modeling and Design (COMAD).