Multiobjective simulation optimization (MOSO) problems are optimization problems with multiple conflicting objectives, where evaluation of at least one of the objectives depends on a black-box numerical code or real-world experiment, which we refer to as a simulation. Whereas an extensive body of research is dedicated to developing new algorithms and methods for solving these and related problems, it is challenging and time-consuming to integrate these techniques into real-world production-ready solvers. This is partly because of the diversity and complexity of modern state-of-the-art MOSO algorithms and methods and partly because of the complexity and specificity of many real-world problems and their corresponding computing environments. The complexity of this problem is only compounded when introducing potentially complex and/or domain-specific surrogate-modeling techniques, problem formulations, design spaces, and data acquisition functions. This paper carefully surveys the current state of the art in MOSO algorithms, techniques, and solvers, as well as problem types and computational environments where MOSO is commonly applied. We then present several key challenges in the design of a parallel multiobjective simulation optimization framework (ParMOO) and how they have been addressed. Finally, we provide two case studies demonstrating how customized ParMOO solvers can be quickly built and deployed to solve real-world MOSO problems. History: Accepted by Ted Ralphs, Area Editor for Software Tools. Funding: This work was supported by US DOE, Office of Science, Advanced Scientific Computing Research, SciDAC Program [Grants DE-AC02-05CH11231 and DE-AC02-06CH11357]. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2023.0250 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2023.0250 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .

Almost all applications stop scaling at some point; those that don't are seldom performant when considering time to solution on anything but aspirational/unicorn resources. Recognizing these tradeoffs as well as greater user functionality in a near-term exascale computing era, we present libEnsemble, a library aimed at particular scalability- and capability-stretching uses. libEnsemble enables running concurrent instances of an application in dynamically allocated ensembles through an extensible Python library. We highlight the structure, execution, and capabilities of the library on leading pre-exascale environments as well as advanced capabilities for exascale environments and beyond.

User-space HPC data services are emerging as an appealing alternative to traditional parallel file systems, because of their ability to be tailored to application needs while eliminating unnecessary overheads incurred by POSIX compliance. The High Energy Physics (HEP) community is progressively turning towards such services to enable high-throughput accesses under heavy concurrency to billions of event data produced by instruments and consumed by subsequent analysis workflows. Such services would benefit from the possibility to be rescaled up and down to adapt to changing workloads, as experimental campaigns progress , in order to optimize resource usage. This paper formalizes rescaling a distributed storage system as a multi objective optimization problem considering three criteria: load balance, data balance, and duration of the rescaling operation. We propose a heuristic for rapidly finding a good approximate solution, while allowing users to weight the criteria as needed. The heuristic is evaluated with Pufferscale, a new rescaling manager for microservice-based distributed storage systems. To validate our approach in a real-world ecosystem, we showcase the use of Pufferscale as a means to enable storage malleability in the HEPnOS storage system for HEP applications.

Scanning-probe x-ray tomography is useful for imaging nanoscale structures of a sample. The image resolution, which can reach down to 10 nm by the increased brightness and coherence of the x-ray optics, however, is highly susceptible to experimental error. Failure to address these errors can lead to a smeared image and, in the worst case, to misinterpretation of the imaged object's structure. In this work, we present a novel optimization-based approach to calibrate a common yet challenging source of experimental error, the drifts of the scanning positions, while simultaneously reconstructing the object. This approach utilizes the coupled and complementary information from different measurements to inform a coherent system between the measurements and the reconstruction. We illustrate the proposed approach on both synthetic and real tomography images and show its superior performance compared without explicit error calibration.

Supervised learning is a promising approach for modeling the performance of applications running on large HPC systems. A key assumption in supervised learning is that the training and testing data are obtained under the same conditions. However, in production HPC systems these conditions might not hold because the conditions of the platform can change over time as a result of hardware degradation, hardware replacement, software upgrade, and configuration updates. These changes could alter the data distribution in a way that affects the accuracy of the predictive performance models and render them less useful; this phenomenon is referred to as concept drift. Ignoring concept drift can lead to suboptimal resource usage and decreased efficiency when those performance models are deployed for tuning and job scheduling in production systems. To address this issue, we propose a concept-drift-aware predictive modeling approach that comprises two components: (1) an online Bayesian changepoint detection method that can automatically identify the location of events that lead to concept drift in near-real time and (2) a moment-matching transformation inspired by transfer learning that converts the training data collected before the drift to be useful for retraining.

Cancer is a complex disease, the understanding and treatment of which are being aided through increases in the volume of collected data and in the scale of deployed computing power. Consequently, there is a growing need for the development of data-driven and, in particular, deep learning methods for various tasks such as cancer diagnosis, detection, prognosis, and prediction. Despite recent successes, however, designing high-performing deep learning models for nonimage and nontext cancer data is a time-consuming, trial-and-error, manual task that requires both cancer domain and deep learning expertise. To that end, we develop a reinforcement-learning-based neural architecture search to automate deep-learning-based predictive model development for a class of representative cancer data. We develop custom building blocks that allow domain experts to incorporate the cancer-data-specific characteristics. We show that our approach discovers deep neural network architectures that have significantly fewer trainable parameters, shorter training time, and accuracy similar to or higher than those of manually designed architectures. We study and demonstrate the scalability of our approach on up to 1,024 Intel Knights Landing nodes of the Theta supercomputer at the Argonne Leadership Computing Facility.

Storage system performance modeling is crucial for efficient use of heterogeneous shared resources on leadership-class computers. Variability in application performance, particularly variability arising from concurrent applications sharing I/O resources, is a major hurdle in the development of accurate performance models. We adopt a deep learning approach based on conditional variational auto encoders (CVAE) for I/O performance modeling, and use it to quantify performance variability. We illustrate our approach using the data collected on Edison, a production supercomputing system at the National Energy Research Scientific Computing Center (NERSC). The CVAE approach is investigated by comparing it to a previously proposed sensitivity-based Gaussian process (GP) model. We find that the CVAE model performs slightly better than the GP model in cases where training and testing data come from different applications, since CVAE can inherently leverage the whole data from multiple applications whereas GP partitions the data and builds separate models for each partition. Hence, the CVAE offers an alternative modeling approach that does not need pre-processing; it has enough flexibility to handle data from a wide variety of applications without changing the inference approach.

Hyperparameters employed by deep learning (DL) methods play a substantial role in the performance and reliability of these methods in practice. Unfortunately, finding performance optimizing hyperparameter settings is a notoriously difficult task. Hyperparameter search methods typically have limited production-strength implementations or do not target scalability within a highly parallel machine, portability across different machines, experimental comparison between different methods, and tighter integration with workflow systems. In this paper, we present DeepHyper, a Python package that provides a common interface for the implementation and study of scalable hyperparameter search methods. It adopts the Balsam workflow system to hide the complexities of running large numbers of hyperparameter configurations in parallel on high-performance computing (HPC) systems. We implement and study asynchronous model-based search methods that consist of sampling a small number of input hyperparameter configurations and progressively fitting surrogate models over the input-output space until exhausting a user-defined budget of evaluations. We evaluate the efficacy of these methods relative to approaches such as random search, genetic algorithms, Bayesian optimization, and hyperband on DL benchmarks on CPU-and GPU-based HPC systems.

Storage resources in high-performance computing are shared across all user applications. Consequently, storage performance can vary markedly, depending not only on an application's workload but also on what other activity is concurrently running across the system. This variability in storage performance is directly reflected in overall execution time variability, thus confounding efforts to predict job performance for scheduling or capacity planning. I/O variability also complicates the seemingly straightforward process of performance measurement when evaluating application optimizations. In this work we present a methodology to measure I/O contention with more rigor than in prior work. We apply statistical techniques to gain insight from application-level statistics and storage-side logging. We examine different correlation metrics for relating system workload to job I/O performance and identify an effective and generally applicable metric for measuring job I/O performance. We further demonstrate that the system-wide monitoring granularity can directly affect the strength of correlation observed. Insufficient granularity and measurements can hide the correlations between application I/O performance and system-wide I/O activity.

Scientific user facilities — particle accelerators, telescopes, colliders, supercomputers, light sources, sequencing facilities, and more — operated by the U.S. Department of Energy (DOE) Office of Science (SC) generate ever increasing volumes of data at unprecedented rates from experiments, observations, and simulations. At the same time there is a growing community of experimentalists that require real-time data analysis feedback, to enable them to steer their complex experimental instruments to optimized scientific outcomes and new discoveries. Recent efforts in DOE-SC have focused on articulating the data-centric challenges and opportunities facing these science communities. Key challenges include difficulties coping with data size, rate, and complexity in the context of both real-time and post-experiment data analysis and interpretation. Solutions will require algorithmic and mathematical advances, as well as hardware and software infrastructures that adequately support data-intensive scientific workloads. This paper presents the summary findings of a workshop held by DOE-SC in September 2015, convened to identify the major challenges and the research that is needed to meet those challenges.

An augmented Lagrangian (AL) can convert a constrained optimization problem into a sequence of simpler (e.g., unconstrained) problems which are then usually solved with local solvers. Recently, surrogate-based Bayesian optimization (BO) sub-solvers have been successfully deployed in the AL framework for a more global search in the presence of inequality constraints; however a drawback was that expected improvement (EI) evaluations relied on Monte Carlo. Here we introduce an alternative slack variable AL, and show that in this formulation the EI may be evaluated with library routines. The slack variables furthermore facilitate equality as well as inequality constraints, and mixtures thereof. We show our new slack "ALBO" compares favorably to the original. Its superiority over conventional alternatives is reinforced on several new mixed constraint examples.

Fluorescence tomographic reconstruction, based on the detection of photons coming from fluorescent emission, can be used for revealing the internal elemental composition of a sample. On the other hand, conventional X-ray transmission tomography can be used for reconstructing the spatial distribution of the absorption coefficient inside a sample. In this work, we integrate both X-ray fluorescence and X-ray transmission data modalities and formulate a nonlinear optimization-based approach for reconstruction of the elemental composition of a given object. This model provides a simultaneous reconstruction of both the quantitative spatial distribution of all elements and the absorption effect in the sample. Mathematically speaking, we show that compared with the single-modality inversion (i.e., the X-ray transmission or fluorescence alone), the joint inversion provides a better-posed problem, which implies a better recovery. Therefore, the challenges in X-ray fluorescence tomography arising mainly from the effects of self-absorption in the sample are partially mitigated. The use of this technique is demonstrated on the reconstruction of several synthetic samples.

Parallel I/O performance depends highly on the interactions among multiple layers of the parallel I/O stack. The most common layers include high-level I/O libraries, MPI-IO middleware, and parallel file system. Each of these layers offers various tunable parameters to control intermediary data transfer points and the final data layout. Due to the interdependencies and the number of combinations of parameters, finding a good set of parameter values for a specific application's I/O pattern is challenging. Recent efforts, such as autotuning with genetic algorithms (GAs) and analytical models, have several limitations. For instance, analytical models fail to capture the dynamic nature of shared supercomputing systems and are application-specific. GA-based tuning requires running many time-consuming experiments for each input size. In this paper, we present a strategy to generate automatically an empirical model for a given application pattern. Using a set of real measurements from running an I/O kernel as training set, we generate a nonlinear regression model. We use this model to predict the top-20 tunable parameter values that give efficient I/O performance and rerun the I/O kernel to select the best set of parameter under the current conditions as tunable parameters for future runs of the same I/O kernel. Using this approach, we demonstrate 6X - 94X speedup over default I/O time for different I/O kernels running on multiple HPC systems. We also evaluate performance by identifying interdependencies among different sets of tunable parameters.

The optimization of parallel I/O has become challenging because of the increasing storage hierarchy, performance variability of shared storage systems, and the number of factors in the hardware and software stacks that impact performance. In this paper, we perform an in-depth study of the complexity involved in I/O autotuning and performance modeling, including the architecture, software stack, and noise. We propose a novel hybrid model combining analytical models for communication and storage operations and black-box models for the performance of the individual operations. The experimental results show that the hybrid approach performs significantly better and shows a higher robustness to noise than state-of-the-art machine learning approaches, at the cost of a higher modeling complexity.

Many factors affect the performance and power characteristics of FPGA designs. Among them are the optimization parameters for synthesis, map, and place-and-route design tools. Choosing the right combination of these parameters can substantially lower power requirements, while still satisfying timing constraints. Finding such an improvement, however, requires significant experimentation by the FPGA designer. Exhaustive search through the parameter space is an automated alternative to experimentation but is intractable because of the large search space and the long build time of each parameter combination. In this paper, we propose a machine-learning-based approach to tune FPGA design parameters. It performs sampling-based reduction of the parameter space and guides the search toward promising parameter configurations. In our experiments, such selective sampling finds parameter configurations that meet the timing constraints and are within 0.2% of the optimal power consumption. Furthermore, these configurations are found in an order of magnitude less time compared with exhaustive search. Such speedups can substantially alleviate bottlenecks in the FPGA design process.

Increased complexity of computer architectures, consideration of power constraints, and expected failure rates of hardware components make the design and analysis of energy-efficient fault-tolerance schemes an increasingly challenging and important task. We develop run-time and study FTI, a multilevel checkpoint library, on an IBM Blue Gene/Q. We show that FTI has a low energy footprint and that, consequently optimal checkpoint-interval values with respect to time and energy are similar.

Various layers of the parallel I/O subsystem offer tunable parameters for improving I/O performance on large-scale computers. However, searching through a large parameter space is challenging. We are working towards an autotuning framework for determining the parallel I/O parameters that can achieve good I/O performance for different data write patterns. In this paper, we characterize parallel I/O and discuss the development of predictive models for use in effectively reducing the parameter space. Applying our technique on tuning an I/O kernel derived from a large-scale simulation code shows that the search time can be reduced from 12 hours to 2 hours, while achieving 54X I/O performance speedup.

Performance models have profound impact on hardware-software codesign, architectural explorations, and performance tuning of scientific applications. Developing algebraic performance models is becoming an increasingly challenging task. In such situations, a statistical surrogate-based performance model, fitted to a small number of input-output points obtained from empirical evaluation on the target machine, provides a range of benefits. Accurate surrogates can emulate the output of the expensive empirical evaluation at new inputs and therefore can be used to test and/or aid search, compiler, and autotuning algorithms. We present an iterative parallel algorithm that builds surrogate performance models for scientific kernels and workloads on single-core and multicore and multinode architectures. We tailor to our unique parallel environment an active learning heuristic popular in the literature on the sequential design of computer experiments in order to identify the code variants whose evaluations have the best potential to improve the surrogate. We use the proposed approach in a number of case studies to illustrate its effectiveness.

We employ recent work on computational noise to obtain near-optimal difference estimates of the derivative of a noisy function. Our analysis relies on a stochastic model of the noise without assuming a specific form of distribution. We use this model to derive theoretical bounds for the errors in the difference estimates and obtain an easily computable difference parameter that is provably near-optimal. Numerical results closely resemble the theory and show that we obtain accurate derivative estimates even when the noisy function is deterministic.

We consider the problem faced by a company that wants to use viral marketing to introduce a new product into a market where a competing product is already being introduced. We assume that consumers will use only one of the two products and will influence their friends in their decision of which product to use. We propose two models for the spread of influence of competing technologies through a social network and consider the influence maximization problem from the follower's perspective. In particular we assume the follower has a fixed budget available that can be used to target a subset of consumers and show that, although it is NP-hard to select the most influential subset to target, it is possible to give an efficient algorithm that is within 63% of optimal. Our computational experiments show that by using knowledge of the social network and the set of consumers targeted by the competitor, the follower may in fact capture a majority of the market by targeting a relatively small set of the right consumers.