SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
This talk provides an overview of the FY22 allocation process for obtaining computing time on the Eagle supercomputer. The overall process is outlined, with specific instructions on completing the request.
The rising number of applications of machine learning and computational statistics in fusion energy research requires flexibility in adopting a growing variety of tools. The Computational Physics School for Fusion Research (CPS-FR) aims at providing young researchers with critical skill sets to deal with modern fusion energy research challenges. The School aims at covering essentials of: Computational Statistics, Machine Learning, Deep Learning and optimization methods, Parallel Programming and HPC. As the first edition of the CPS-FR just concluded, this report highlights its main results and summarizes its contents.
Abstract Background The Basic Local Alignment Search Tool (BLAST) is a suite of commonly used algorithms for identifying matches between biological sequences. The user supplies a database file and query file of sequences for BLAST to find identical sequences between the two. The typical millions of database and query sequences make BLAST computationally challenging but also well suited for parallelization on high-performance computing clusters. The efficacy of parallelization depends on the data partitioning, where the optimal data partitioning relies on an accurate performance model. In previous studies, a BLAST job was sped up by 27 times by partitioning the database and query among thousands of processor nodes. However, the optimality of the partitioning method was not studied. Unlike BLAST performance models proposed in the literature that usually have problem size and hardware configuration as the only variables, the execution time of a BLAST job is a function of database size, query size, and hardware capability. In this work, the nucleotide BLAST application BLASTN was profiled using three methods: shell-level profiling with the Unix “time” command, code-level profiling with the built-in “profiler” module, and system-level profiling with the Unix “gprof” program. The runtimes were measured for six node types, using six different database files and 15 query files, on a heterogeneous HPC cluster with 500+ nodes. The empirical measurement data were fitted with quadratic functions to develop performance models that were used to guide the data parallelization for BLASTN jobs. Results Profiling results showed that BLASTN contains more than 34,500 different functions, but a single function, RunMTBySplitDB, takes 99.12% of the total runtime. Among its 53 child functions, five core functions were identified to make up 92.12% of the overall BLASTN runtime. Based on the performance models, static load balancing algorithms can be applied to the BLASTN input data to minimize the runtime of the longest job on an HPC cluster. Four test cases being run on homogeneous and heterogeneous clusters were tested. Experiment results showed that the runtime can be reduced by 81% on a homogeneous cluster and by 20% on a heterogeneous cluster by re-distributing the workload. Discussion Optimal data partitioning can improve BLASTN’s overall runtime 5.4-fold in comparison with dividing the database and query into the same number of fragments. The proposed methodology can be used in the other applications in the BLAST+ suite or any other application as long as source code is available.
Under this grant, three significant software packages were developed or improved, all with the goal of improving the ease-of-use of HPC libraries. The first component is a Python package, named DistArray (originally named Odin), that provides a high-level interface to distributed array computing. This interface is based on the popular and widely used NumPy package and is integrated with the IPython project for enhanced interactive parallel distributed computing. The second Python package is the Distributed Array Protocol (DAP) that enables separate distributed array libraries to share arrays efficiently without copying or sending messages. If a distributed array library supports the DAP, it is then automatically able to communicate with any other library that also supports the protocol. This protocol allows DistArray to communicate with the Trilinos library via PyTrilinos, which was also enhanced during this project. A third package, PyTrilinos, was extended to support distributed structured arrays (in addition to the unstructured arrays of its original design), allow more flexible distributed arrays (i.e., the restriction to double precision data was lifted), and implement the DAP. DAP support includes both exporting the protocol so that external packages can use distributed Trilinos data structures, and importing the protocol so that PyTrilinos can work with distributed data from external packages.
Overview: Anonymized job-level records from the Kestrel HPC system at the National Laboratory of the Rockies (NLR). Each record represents a Slurm batch job with scheduling metadata, resource requests, utilization, energy estimates, and efficiency metrics. Sensitive fields (user, account, job name, submit line, working directory, submit script, and job type) are replaced with 7-character cryptographic hashes. System & Timeframe: Kestrel is located at the NLR campus. Standard compute nodes have 104 cores and 256 GB RAM; bigmem nodes have 2,000 GB. GPU nodes (gpu-h100 partition) use NVIDIA H100 GPUs. Data covers jobs submitted August 2023 through December 2025. Funding provided by the U.S. Department of Energy, EERE. Files: esif.hpc.kestrel.job-anon.zip — Anonymized job records (Hive-partitioned Parquet) datacard.md — Full dataset documentation ~11 million rows, 50 variables. Readable with PyArrow, pandas, DuckDB, Apache Spark, or any Parquet-compatible tool. Data Collection: Jobs collected via sacct with timezone-aware export (SLURM_TIME_FORMAT="%Y-%m-%dT%H:%M:%S%z"), loaded into PostgreSQL. Calculated columns updated via database triggers and batch functions. All timestamps use timestamptz and correctly handle DST transitions. Preprocessing: Anonymization of name, user, account, submit_line, work_dir, submit_script, and job_type via 7-char hex hashes Derived columns: queue_wait, cpu_eff, max/min/avg_mem_eff, energy estimates Simplified job state mapping (e.g., "CANCELLED by 132357" → "CANCELLED") Boolean flags: python_job, reframe_job Temporal decomposition: year, month, day, day_of_week, hour, minute from submit_time Shared node tracking: shared_job_count, nodes_shared, jobs_shared Key Variables: Scheduling: job_id, partition, state_simple, submit_time, start_time, end_time, queue_wait Resources: nodes_req/used, processors_req/used, memory_req, wallclock_req/used, gpus_requested Efficiency: cpu_eff, max/min/avg_mem_eff Energy: cpu_energy_tdp_estimated_max/used_watt_hours, consumed_energy_raw_joules, consumed_energy_raw_watt_hours Sharing: shared_job_count, nodes_shared, jobs_shared Partitions: short, standard, debug, gpu-h100 Job States: CANCELLED, COMPLETED, FAILED, PENDING, RUNNING QoS Levels: normal, high Important Notes: Timestamps include timezone offsets; DST transitions are handled correctly, though adding intervals across DST boundaries requires offset adjustment shared_job_count reflects physical node co-residency, not use of the shared partition Job step records and raw Slurm JSONB fields are excluded Do not attempt to re-identify individuals from hashed fields
An open-source program named VeloxChem has been developed for the calculation of electronic real and complex linear response functions at the levels of Hartree–Fock and Kohn–Sham density functional theories. With an object-oriented program structure written in a Python/C++ layered fashion, VeloxChem enables time-efficient prototyping of novel scientific approaches without sacrificing computational efficiency, so that molecular systems involving up to and beyond 500 second-row atoms (or some 10,000 contracted and in part diffuse Gaussian basis functions) can be routinely addressed. In addition, VeloxChem is equipped with a polarizable embedding scheme for the treatment of the classical electrostatic interactions with an environment that in turn is modeled by atomic site charges and polarizabilities. The underlying hybrid message passing interface (MPI)/open multiprocessing (OpenMP) parallelization scheme makes VeloxChem suitable for execution in high-performance computing cluster environments, showing even slightly beyond linear scaling for the Fock matrix construction with use of up to 16,384 central processing unit (CPU) cores. An efficient—with respect to convergence rate and overall computational cost—multifrequency/gradient complex linear response equation solver enables calculations not only of conventional spectra, such as visible/ultraviolet/X-ray electronic absorption and circular dichroism spectra, but also time-resolved linear response signals as due to ultra-short weak laser pulses.
Ensembles of simulations are employed to estimate the statistics of possible future states of a system, and are widely used in important applications such as climate change and biological modeling. Ensembles of runs can naturally be executed in parallel. However, when the CPU times of individual simulations vary considerably, a simple strategy of assigning an equal number of tasks per processor can lead to serious work imbalances and low parallel efficiency. This paper presents a new probabilistic framework to analyze the performance of dynamic load balancing algorithms for ensembles of simulations where many tasks are mapped onto each processor, and where the individual compute times vary considerably among tasks. Four load balancing strategies are discussed: most-dividing, all-redistribution, random-polling, and neighbor-redistribution. Simulation results with a stochastic budding yeast cell cycle model are consistent with the theoretical analysis. It is especially significant that there is a provable global decrease in load imbalance for the local rebalancing algorithms due to scalability concerns for the global rebalancing algorithms. The overall simulation time is reduced by up to 25 %, and the total processor idle time by 85 %.
The geological disposal of nuclear waste is based on the multi-barrier concept, comprising various engineered and natural barriers, to confine the radioactive waste and isolate it from the biosphere. Some of the planned repositories for high-level nuclear waste will be hosted in fractured crystalline rock formations. The potential of these formations to act as natural transport barriers is related to two coupled processes: diffusion into the rock matrix and sorption onto the mineral surfaces available in the rock matrix. Different in situ and laboratory experiments have pointed out the ubiquitous heterogeneous nature of the rock matrix: mineral surfaces and pore space are distributed in complex microstructures and their distribution is far from being homogeneous (as typically assumed by Darcy-scale coarse reactive transport models). We use a synthetically generated fracture–matrix system to assess the implications of grain-scale physical and mineralogical heterogeneity on cesium transport and retention. The resulting grain-scale reactive transport model is solved using high-performance computing technologies, and the results are compared with those derived from two alternative models, denoted as upscaled models, where mineral abundance is averaged over the matrix volume. In the grain-scale model, the penetration of cesium into the matrix is faster and the penetration front is uneven and finger-shaped. The analysis of the cesium breakthrough curves computed at two different points in the fracture shows that the upscaled models provide later first-arrival time estimates compared to the grain-scale model. The breakthrough curves computed with the three models converge at late times. These results suggest that spatially averaged upscaled parameters of sorption site distribution can be used to predict the late-time behavior of breakthrough curves but could be inadequate to simulate the early behavior.
Models that address policy-maker concerns about multi-scale effects of food and bioenergy production systems are computationally demanding. We integrated the message passing interface algorithm into the process-based EPIC model to accelerate computation of ecosystem effects. Simulation performance was further enhanced by applying the Vampir framework. When this enhanced mpi_EPIC model was tested, total execution time for a global 30-year simulation of a switchgrass cropping system was shortened to less than 0.5 hours on a supercomputer. The results illustrate that mpi_EPIC using parallel design can balance simulation workloads and facilitate large-scale, high-resolution analysis of agricultural production systems, management alternatives and environmental effects.
Compressible reacting flows may display sharp spatial variation related to shocks, contact discontinuities or reactive zones embedded within relatively smooth regions. The presence of such phenomena emphasizes the relevance of shock-capturing schemes such as the weighted essentially non-oscillatory (WENO) scheme as an essential ingredient of the numerical solver. However, these schemes are complex and have more computational cost than the simple high-order compact or non-compact schemes. In this paper, we present the implementation of a seventh-order, minimally-dissipative mapped WENO (WENO7M) scheme in a newly developed direct numerical simulation (DNS) code called KAUST Adaptive Reactive Flows Solver (KARFS). In order to make efficient use of the computer resources and reduce the solution time, without compromising the resolution requirement, the WENO routines are accelerated via graphics processing unit (GPU) computation. The performance characteristics and scalability of the code are studied using different grid sizes and block decomposition. Furthermore, the performance portability of KARFS is demonstrated on a variety of architectures including NVIDIA Tesla P100 GPUs and NVIDIA Kepler K20X GPUs. In addition, the capability and potential of the newly implemented WENO7M scheme in KARFS to perform DNS of compressible flows is also demonstrated with model problems involving shocks, isotropic turbulence, detonations and flame propagation into a stratified mixture with complex chemical kinetics.
Software simulators play a critical role in the development of new algorithms and system architectures in any field of engineering. Neuromorphic computing, which has shown potential in building brain-inspired energy-efficient hardware, suffers a slow-down in the development cycle due to a lack of flexible and easy-to-use simulators of either neuromorphic hardware itself or of spiking neural networks (SNNs), the type of neural network computation executed on most neuromorphic systems. While there are several openly available neuromorphic or SNN simulation packages developed by a variety of research groups, they have mostly targeted computational neuroscience simulations, and only a few have targeted small-scale machine learning tasks with SNNs. Evaluations or comparisons of these simulators have often targeted computational neuroscience-style workloads. In this work, we seek to evaluate the performance of several publicly available SNN simulators with respect to non-computational neuroscience workloads, in terms of speed, flexibility, and scalability. We evaluate the performance of the NEST, Brian2, Brian2GeNN, BindsNET and Nengo packages under a common front-end neuromorphic framework. Our evaluation tasks include a variety of different network architectures and workload types to mimic the computation common in different algorithms, including feed-forward network inference, genetic algorithms, and reservoir computing. We also study the scalability of each of these simulators when running on different computing hardware, from single core CPU workstations to multi-node supercomputers. Our results show that the BindsNET simulator has the best speed and scalability for most of the SNN workloads (sparse, dense, and layered SNN architectures) on a single core CPU. However, when comparing the simulators leveraging the GPU capabilities, Brian2GeNN outperforms the others for these workloads in terms of scalability. NEST performs the best for small sparse networks and is also the most flexible simulator in terms of reconfiguration capability NEST shows a speedup of at least 2x compared to the other packages when running evolutionary algorithms for SNNs. The multi-node and multi-thread capabilities of NEST show at least 2x speedup compared to the rest of the simulators (single core CPU or GPU based simulators) for large and sparse networks. We conclude our work by providing a set of recommendations on the suitability of employing these simulators for different tasks and scales of operations. We also present the characteristics for a future generic ideal SNN simulator for different neuromorphic computing workloads.
Not Available
We develop and implement a multifrequency spectroscopy and spectroscopic imaging mode, referred to as general dynamic mode (GDM), that captures the complete spatially- and stimulus dependent information on nonlinear cantilever dynamics in scanning probe microscopy (SPM). GDM acquires the cantilever response including harmonics and mode mixing products across the entire broadband cantilever spectrum as a function of excitation frequency. GDM spectra substitute the classical measurements in SPM, e.g. amplitude and phase in lock-in detection. Here, GDM is used to investigate the response of a purely capacitively driven cantilever. We use information theory techniques to mine the data and verify the findings with governing equations and classical lock-in based approaches. We explore the dependence of the cantilever dynamics on the tip–sample distance, AC and DC driving bias. This approach can be applied to investigate the dynamic behavior of other systems within and beyond dynamic SPM. In conclusion, GDM is expected to be useful for separating the contribution of different physical phenomena in the cantilever response and understanding the role of cantilever dynamics in dynamic AFM techniques.
Because large unstructured datasets is important for many science domains, distributed graph analytics is critical to many scientists. Unfortunately, obtaining scaling and performance for irregular communication is challenging because contemporary network interconnects are primarily designed to maximize bandwidths of fixed-neighborhoods large-message exchanges (e.g., stencils). Although there is no consensus on the “best” network topologies for irregular communication, unstructured graph-based interconnects can be more suitable. We analyze three popular graph workloads – clustering, pattern enumeration, and traversal — on comparable networks (in terms of resources and costs) constructed from Jellyfish Random Regular, Dragonfly and Fat tree topologies, varying the routing algorithms. Using packet-level simulations, we demonstrate up to 60% improvement in communication time with Jellyfish due to diversity of the short paths between arbitrary endpoints, which can reduce overall network stalls and congestion.
HPC (high-performance computing) applications usually show bursty I/O behaviors. In order to expedite the applications, permanent storage systems are usually provisioned to serve such I/O bursts. Approaching the era of exascale computing, non-volatile RAM is introduced as burst buffers, to absorb the bursty bulk data and relax the I/O provisioning requirement of the permanent storage systems. However, without judiciously draining the burst buffers, I/O bursts are passed down to the underlying storage systems, which causes severe I/O contention issues.In order to minimize the I/O provisioning requirement and resolve the issues caused by I/O bursts, we propose a proactive draining scheme to manage the draining process of distributed node-local burst buffers. In addition, we develop an I/O provisioning model to predict the minimized I/O provisioning requirement for permanent storage systems. Evaluation results show that applying the proactive draining scheme largely relaxes the I/O provisioning requirement while preserving the I/O performance of underlying storage systems.
Driven by the trends of increasing core-count and bandwidth-wall problem, the size of last level caches (LLCs) has greatly increased and hence, the researchers have explored non-volatile memories (NVMs) which provide high density and consume low-leakage power. Since NVMs have low write-endurance and the existing cache management policies are write variation-unaware, effective wear-leveling techniques are required for achieving reasonable cache lifetimes using NVMs. We present EqualWrites, a technique for mitigating intra-set write variation. In this paper, our technique works by recording the number of writes on a block and changing the cache-block location of a hot data-item to redirect the future writes to a cold block to achieve wear-leveling. Simulation experiments have been performed using an x86-64 simulator and benchmarks from SPEC06 and HPC (high-performance computing) field. The results show that for single, dual and quad-core system configurations, EqualWrites improves cache lifetime by 6.31X, 8.74X and 10.54X, respectively. In addition, its implementation overhead is very small and it provides larger improvement in lifetime than three other intra-set wear-leveling techniques and a cache replacement policy.
Here, the collaborative efforts of large communities in science experiments, often comprising thousands of global members, reflect a monumental commitment to exploration and discovery. Recently, advanced and complex data processing has gained increasing importance in science experiments. Data processing workflows typically consist of multiple intricate steps, and the precise specification of resource requirements is crucial for each step to allocate optimal resources for effective processing. Estimating resource requirements in advance is challenging due to a wide range of analysis scenarios, varying skill levels among community members, and the continuously increasing spectrum of computing options. One practical approach to mitigate these challenges involves initially processing a subset of each step to measure precise resource utilization from actual processing profiles before completing the entire step. While this two-staged approach enables processing on optimal resources for most of the workflow, it has drawbacks such as initial inaccuracies leading to potential failures and suboptimal resource usage, along with overhead from waiting for initial processing completion, which is critical for fast-turnaround analyses. In this context, our study introduces a novel pipeline of machine learning models within a comprehensive workflow management system, the Production and Distributed Analysis (PanDA) system. These models employ advanced machine learning techniques to predict key resource requirements, overcoming challenges posed by limited upfront knowledge of characteristics at each step. Accurate forecasts of resource requirements enable informed and proactive decision-making in workflow management, enhancing the efficiency of handling diverse, complex workflows across heterogeneous resources.
High-performance computing systems (HPC) provide powerful capabilities for modeling, simulation, and data analytics for a broad class of computational problems. They enable extreme performance of the order of quadrillion floating-point arithmetic calculations per second by aggregating the power of millions of compute, memory, networking and storage components. With the rapidly growing scale and complexity of HPC systems for achieving even greater performance, ensuring their reliable operation in the face of system degradations and failures is a critical challenge. System fault events often lead the scientific applications to produce incorrect results, or may even cause their untimely termination. The sheer number of components in modern extreme-scale HPC systems and the complex interactions and dependencies among the hardware and software components, the applications, and the physical environment makes the design of practical solutions that support fault resilience a complex undertaking. To manage this complexity, we developed a methodology for designing HPC resilience solutions using design patterns. We codified the well-known techniques for handling faults, errors and failures that have been devised, applied and improved upon over the past three decades in the form of design patterns. In this paper, we present a pattern language to enable a structured approach to the development of HPC resilience solutions. The pattern language reveals the relations among the resilience patterns and provides the means to explore alternative techniques for handling a specific fault model that may have different efficiency and complexity characteristics. Using the pattern language enables the design and implementation of comprehensive resilience solutions as a set of interconnected resilience patterns that can be instantiated across layers of the system stack.