Search

Data management is becoming increasingly important in dealing with the large amounts of data produced by large-scale scientific simulations and instruments. 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 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 70X, and the proposed compression method can improve compression ratio by up to 2X compared with other state-of-the-art error-bounded lossy compressors under the same level of data distortion.

A growing disparity between supercomputer computation speeds and I/O rates means that it is rapidly becoming infeasible to analyze supercomputer application output only after that output has been written to a file system. Instead, data-generating applications must run concurrently with data reduction and/or analysis operations, with which they exchange information via high-speed methods such as interprocess communications. The resulting parallel computing motif, online data analysis and reduction (ODAR), has important implications for both application and HPC systems design. Here we introduce the ODAR motif and its co-design concerns, describe a co-design process for identifying and addressing those concerns, present tools that assist in the co-design process, and present case studies to illustrate the use of the process and tools in practical settings.

We present the Exascale Framework for High Fidelity coupled Simulations (EFFIS), a workflow and code coupling framework developed as part of the Whole Device Modeling Application (WDMApp) in the Exascale Computing Project. EFFIS consists of a library, command line utilities, and a collection of run-time daemons. Together, these software products enable users to easily compose and execute workflows that include: strong or weak coupling, in situ (or offline) analysis/visualization/monitoring, command-and-control actions, remote dashboard integration, and more. We describe WDMApp physics coupling cases and computer science requirements that motivate the design of the EFFIS framework. Furthermore, we explain the essential enabling technology that EFFIS leverages: ADIOS for performant data movement, PerfStubs/TAU for performance monitoring, and an advanced COUPLER for transforming coupling data from its native format to the representation needed by another application. Finally, we demonstrate EFFIS using coupled multi-simulation WDMApp workflows and exemplify how the framework supports the project’s needs. We show that EFFIS and its associated services for data movement, visualization, and performance collection does not introduce appreciable overhead to the WDMApp workflow and that the resource-dominant application’s idle time while waiting for data is minimal.

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%., 12 pages, 15 figures, accepted by IEEE Transactions on Parallel and Distributed Systems

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.

Previous work on multilevel techniques for compression and reduction of scientific data is extended to the case of data given on unstructured meshes in two and three dimensions. The centerpiece of ...

Simulation based scientific applications generate increasingly large amounts of data on high-performance computing (HPC) systems. To allow data to be stored and analyzed efficiently, data compression is often utilized to reduce the volume and velocity of data. However, a question often raised by domain scientists is the level of compression that can be expected so that they can make more informed decisions, balancing between accuracy and performance. In this letter, we propose a deep neural network based approach for estimating the compressibility of scientific data. To train the neural network, we build both general features as well as compressor-specific features so that the characteristics of both data and lossy compressors are captured in training. Our approach is demonstrated to outperform a prior analytical model as well as a sampling based approach in the case of a biased estimation, i.e., for SZ. However, for the unbiased estimation (i.e., ZFP), the sampling based approach yields the best accuracy, despite the high overhead involved in sampling the target dataset.

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.

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 naive 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.

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.

The global nature of the ITER project along with its projected approximately petabyte-per-day data generation presents not only a unique challenge but also an opportunity for the fusion community to rethink, optimize, and enhance our scientific discovery process. Recognizing this, collaborative research with computational scientists was undertaken over the past several years to create a framework for large-scale data movement across wide-area networks to enable global near-real-time analysis of fusion data. This would broaden the available computational resources for analysis/simulation and increase the number of researchers actively participating in experiments. An official demonstration of this framework for fast, large data transfer and real-time analysis was carried out between the KSTAR tokamak in Daejeon, Korea, and Princeton Plasma Physics Laboratory (PPPL) in Princeton, New Jersey. Streaming large data transfer, with near-real-time movie creation and analysis of the KSTAR electron cyclotron emission imaging data, was performed using the Adaptable Input Output (I/O) System (ADIOS) framework, and comparisons were made at PPPL with simulation results from the XGC1 code. These demonstrations were made possible utilizing an optimized network configuration at PPPL, which achieved over 8.8 Gbps (88% utilization) in throughput tests from the National Fusion Research Institute to PPPL. This demonstration showed the feasibility for large-scale data analysis of KSTAR data and provides a nascent framework to enable use of globally distributed computational and personnel resources in pursuit of scientific knowledge from the ITER experiment.

We are developing the Delta framework that aims to tackle big-data challenges specific to fusion energy sciences. Delta can be used to connect fusion experiments to remote supercomputers. Streaming measurements to distributed compute resources allows to automatically perform high-dimensional data analysis on a cadence that exceeds experimental schedules. Making data analysis results available before the next experiments allows scientists to make more informed decisions about configuration of upcoming experiments. Here we describe how Delta uses database and virtualization facilities, as well as high-performance computing, at the National Energy Research Compute Center to offer a vertically integrated near real-time data analysis and visualization. We also report on ongoing efforts to port the data analysis part of Delta to graphical processing units, which show a reduction of the analysis wall-time for a benchmark workflow by about 35% when compared to a serial implementation.

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.

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.

The term “in situ processing” has evolved over the last decade to mean both a specific strategy for visualizing and analyzing data and an umbrella term for a processing paradigm. The resulting confusion makes it difficult for visualization and analysis scientists to communicate with each other and with their stakeholders. To address this problem, a group of over 50 experts convened with the goal of standardizing terminology. This paper summarizes their findings and proposes a new terminology for describing in situ systems. An important finding from this group was that in situ systems are best described via multiple, distinct axes: integration type, proximity, access, division of execution, operation controls, and output type. This paper discusses these axes, evaluates existing systems within the axes, and explores how currently used terms relate to the axes.

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 short paper considers time-to-solution for two in situ visualization paradigms: in-line and in-transit. It is a follow-on work to two previous studies. The first study [10] considered time-to-solution (wall clock time) and total cost (total node seconds incurred) for a single visualization algorithm (isosurfacing). The second study [11] considered only total cost and added a second algorithm (volume rendering). This short paper completes the evaluation, considering time-to-solution for both algorithms. In particular, it extends the first study by adding additional insights from including a second algorithm at larger scale and by doing more extended and formal analysis regarding time-to-solution. Further, it complements the second study as the best in situ configuration to choose can vary when considering time-to-solution over cost. It also makes use of the same data corpus used in the second study, although that data corpus has been refactored with time-to-solution in mind.

The traditional model of having simulations write data to disk for offline analysis can be prohibitively expensive on computers with limited storage capacity or I/O bandwidth. In situ data analysis has emerged as a necessary paradigm to address this issue and is expected to play an important role in exascale computing. We demonstrate the various aspects and challenges involved in setting up a comprehensive in situ data analysis pipeline that consists of a simulation coupled with compression and feature tracking routines, a framework for assessing compression quality, a middleware library for I/O and data management, and a workflow tool for composing and running the pipeline. We perform studies of compression mechanisms and parameters on two supercomputers, Summit at Oak Ridge National Laboratory and Theta at Argonne National Laboratory, for two example application pipelines. We show that the optimal choice of compression parameters varies with data, time, and analysis, and that periodic retuning of the in situ pipeline can improve compression quality. Finally, we discuss our perspective on the wider adoption of in situ data analysis and management practices and technologies in the HPC community.

Checkpoint/Restart (C/R) is widely used to provide fault tolerance on High-Performance Computing (HPC) systems. However, Parallel File System (PFS) overhead and failure uncertainty cause significant application overhead. This paper develops an adaptive multi-level C/R model that incorporates a failure prediction and analysis model, which orchestrates failure prediction, checkpointing, checkpoint frequency, and proactive live migration along with the additional benefit of Burst Buffers (BB). It effectively reduces the overheads due to failures, checkpointing, and recovery. Simulation results for the Summit supercomputer yield a reduction of ~20%-86% in application overhead due to BBs, orchestrated failure prediction, and migration. We also observe a ~29% decrease in checkpoint writes to BBs, which can increase the longevity of the BB storage devices.

Although many compression algorithms are focused on preserving pointwise values of the data, application scientists are generally more concerned with derived quantities. Equally well, the user may ...

We develop a technique for multigrid adaptive reduction of data (MGARD). Special attention is given to the case of tensor product grids, where our approach permits the use of nonuniformly spaced gr...

Upcoming exascale applications could introduce significant data management challenges due to their large sizes, dynamic work distribution, and involvement of accelerators such as graphical processing units, GPUs. In this work, we explore the performance of reading and writing operations involving one such scientific application on two different supercomputers. Our tests showed that the Adaptable Input and Output System, ADIOS, was able to achieve speeds over 1TB/s, a significant fraction of the peak I/O performance on Summit. We also demonstrated the querying functionality in ADIOS could effectively support common selective data analysis operations, such as conditional histograms. In tests, this query mechanism was able to reduce the execution time by a factor of five. More importantly, ADIOS data management framework allows us to achieve these performance improvements with only a minimal amount of coding effort.

We present a multilevel technique for the compression and reduction of univariate data and give an optimal complexity algorithm for its implementation. A hierarchical scheme offers the flexibility to produce multiple levels of partial decompression of the data so that each user can work with a reduced representation that requires minimal storage whilst achieving the required level of tolerance. The algorithm is applied to the case of turbulence modelling in which the datasets are traditionally not only extremely large but inherently non-smooth and, as such, rather resistant to compression. We decompress the data for a range of relative errors, carry out the usual analysis procedures for turbulent data, and compare the results of the analysis on the reduced datasets to the results that would be obtained on the full dataset. The results obtained demonstrate the promise of multilevel compression techniques for the reduction of data arising from large scale simulations of complex phenomena such as turbulence modelling.

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 in the current HPC software ecosystem. To that end, this paper develops SIRIUS, a progressive JPEG-like data management scheme for storing and analyzing big scientific data. It co-designs 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, using either topological or uniform decimation, 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, SIRIUS provides a paradigm shift towards elastic data analytics and enables end users to make trade-offs between analysis speed and accuracy on-the-fly. This paper further develops algorithms to preserve statistics for data decimation, a common requirement for reducing data. We assess the impact of SIRIUS on unstructured triangular meshes, a pervasive data model used in scientific simulations. In particular, we evaluate two realistic use cases: the blob detection in fusion and high-pressure area extraction in computational fluid dynamics.

Interactive evolutionary algorithms have been applied to personalized search, in which less user fatigue and efficient search are pursued. Motivated by this, we present a fast interactive estimation of distribution algorithm (IEDA) by using the domain knowledge of personalized search. We first induce a Bayesian model to describe the distribution of the new user’s preference on the variables from the social knowledge of personalized search. Then we employ the model to enhance the performance of IEDA in two aspects, that is: 1) dramatically reducing the initial huge space to a preferred subspace and 2) generating the individuals of estimation of distribution algorithm(EDA) by using it as a probabilistic model. The Bayesian model is updated along with the implementation of the EDA. To effectively evaluate individuals, we further present a method to quantitatively express the preference of the user based on the human-computer interactions and train a radial basis function neural network as the fitness surrogate. The proposed algorithm is applied to a laptop search, and its superiorities in alleviating user fatigue and speeding up the search procedure are empirically demonstrated.

ML/AI techniques, particularly based on deep learning, will increasingly be used to accelerate scientific discovery for fusion experiment and simulation. Fusion energy devices have many disparate diagnostic instruments, capturing a broad range of interacting physics phenomena over multiple time and spatial scales. Also, fusion experiments are increasingly built to run longer pulses, with a goal of eventually running a reactor continuously. The confluence of these facts leads to large, complex datasets with phenomena manifest over long sequences. A key challenge is enabling scientists/engineers to utilize these datasets, for example to automatically catalog events of interest, predict the onset of phenomena such as tokamak disruptions, and enable comparisons to models/simulation. Given the size, multiple modalities, and multi-scale nature of fusion data, deep learning models are attractive, but at these scales requires utilizing HPC resources. Many ML/AI techniques not fully utilized now will demand even more HPC resources, such as self-supervised learning to help fusion scientists create AI models with less labelled data, and advanced sequence models which use less GPU memory at the expense of increased compute. Additionally, deep learning models will enable faster, more in-depth analysis than previously available, such as extracting physics model parameters from data using conditional variational autoencoders, instead of slower techniques such as Markov chain Monte Carlo (MCMC). Comparison to simulation will also be enhanced through direct acceleration of simulation kernels using deep learning. These ML/AI techniques will give fusion scientists faster results, allowing more efficient machine use, and faster scientific discovery.

One of the primary challenges facing scientists is extracting understanding from the large amounts of data produced by simulations, experiments, and observational facilities. The use of data across the entire lifetime ranging from real-time to post-hoc analysis is complex and varied, typically requiring a collaborative effort across multiple teams of scientists. Over time, three sets of tools have emerged: one set for analysis, another for visualization, and a final set for orchestrating the tasks. This trifurcated tool set often results in the manual assembly of analysis and visualization workflows, which are one-off solutions that are often fragile and difficult to generalize. To address these challenges, we propose a serviced-based paradigm and a set of abstractions to guide its design. These abstractions allow for the creation of services that can access and interpret data, and enable interoperability for intelligent scheduling of workflow systems. This work results from a codesign process over analysis, visualization, and workflow tools to provide the flexibility required for production use. Finally, this paper describes a forward-looking research and development plan that centers on the concept of visualization and analysis technology as reusable services, and also describes several real-world use cases that implement these concepts.

Fusion energy experiments and simulations provide critical information needed to plan future fusion reactors. As next-generation devices like ITER move toward long-pulse experiments, analyses, including AI and ML, should be performed in a wide range of time and computing constraints, from near-real-time constraints, between-shot analysis, and to campaign-wide long-term analysis. However, the data volume, velocity, and variety make it extremely challenging for analyses using only local computational resources. Researchers need the ability to compose and execute workflows spanning edge resources to large-scale high-performance computing facilities.

We analyze the opportunities for in-transit visualization to provide cost savings compared to in-line visualization. We begin by developing a cost model that includes factors related to both in-line and in-transit which allows comparisons to be made between the two methods. We then run a series of studies to create a corpus of data for our model. We run two different visualization algorithms, one that is computation heavy and one that is communication heavy with concurrencies up to 32, 768 cores. Our primary results are in exploring the cost model within the context of our corpus. Our findings show that in-transit consistently achieves significant cost efficiencies by running visualization algorithms at lower concurrency, and that in many cases these efficiencies are enough to offset other costs (transfer, blocking, and additional nodes) to be cost effective overall. Finally, this work informs future studies, which can focus on choosing ideal configurations for in-transit processing that can consistently achieve cost efficiencies.

The cost of I/O is a significant challenge on current supercomputers, and the trend is likely to continue into the foreseeable future. This challenge is amplified in scientific visualization because of the requirement to consume large amounts of data before processing can begin. Lossy compression has become an important technique in reducing the cost of performing I/O. In this paper we consider the implications of using compressed data for visualization within a scientific workflow. We use visualization operations on simulation data that is reduced using three different state-of-the-art compression techniques. We study the storage efficiency and preservation of visualization features on the resulting compressed data, and draw comparisons between the three techniques used. Our contributions can help inform both scientists and researchers in the use and design of compression techniques for preservation of important visualization details.

In this paper we discuss our design of a toolset for automating performance studies of composed HPC applications that perform online data reduction and analysis. We describe Cheetah, a new framework for performing parametric studies on coupled applications. Cheetah facilitates understanding the impact of various factors such as process placement, synchronicity of algorithms, and storage vs. compute requirements for online analysis of large data. Ultimately, we aim to create a catalog of performance results that can help scientists understand tradeoffs when designing next-generation simulations that make use of online processing techniques. We illustrate the design choices of Cheetah by using a reaction-diffusion simulation (Gray-Scott) paired with an analysis application to demonstrate initial results of fine-grained process placement on Summit, a pre-exascale supercomputer at Oak Ridge National Laboratory.

Complex workflows consisting of multiple simulation and analysis codes running concurrently through in-memory coupling is becoming popular due to inherent advantages in online management of large-scale data, resilience, and the code development process. However, orchestrating such a multi-application workflow to efficiently utilize resources on a heterogeneous architecture is challenging. In this paper, we present our results with running the Fusion Whole Device Modeling benchmark workflow on Summit, a pre-exascale supercomputer at Oak Ridge National Laboratory. We explore various resource distribution and process placement mechanisms, including sharing compute nodes between processes from separate applications. We show that fine-grained process placement can have a significant impact towards efficient utilization of the compute power of a node on Summit, and conclude that sophisticated tools for performing co-design studies of multi-application workflows can play an important role towards efficient orchestration of such workflows.

Part of the promise of exascale computing and the next generation of scientific simulation codes is the ability to bring together time and spatial scales that have traditionally been treated separately. This enables creating complex coupled simulations and in situ analysis pipelines, encompassing such things as "whole device" fusion models or the simulation of cities from sewers to rooftops. Unfortunately, the HPC analysis tools that have been built up over the preceding decades are ill suited to the debugging and performance analysis of such computational ensembles. In this paper, we present a new vision for performance measurement and understanding of HPC codes, MonitoringAnalytics (MONA). MONA is designed to be a flexible, high performance monitoring infrastructure that can perform monitoring analysis in place or in transit by embedding analytics and characterization directly into the data stream, without relying upon delivering all monitoring information to a central database for post-processing. It addresses the trade-offs between the prohibitively expensive capture of all performance characteristics and not capturing enough to detect the features of interest. We demonstrate several uses of MONA; capturing and indexing multi-executable performance profiles to enable later processing, extraction of performance primitives to enable the generation of customizable benchmarks and performance skeletons, and extracting communication and application behaviors to enable better control and placement for the current and future runs of the science ensemble. Relevant performance information based on a system for MONA built from ADIOS and SOSflow technologies is provided for DOE science applications and leadership machines.