Search NASA⌕ Search

SEARCH · Search NASA

Results for “Workload”

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.

At least 523 records · Page 29

Detecting Anomalous Computation with RNNs on GPU-Accelerated HPC Machines

This paper presents a workload classification framework that discriminates illicit computation from authorized workloads on GPU-accelerated HPC systems. As such systems become more and more powerful, they are exploited by attackers to run malicious and for-profit programs that typically require extremely high computing ability to be successful. Our classification framework leverages the distinctive signatures between illicit and authorized workloads, and explore machine learning methods to learn the workloads and classify them. The framework uses lightweight, non-intrusive workload profiling to collect model input data, and explores multiple machine learning methods, particularly recurrent neural network (RNN) that is suitable for online anomalous workload detection. Evaluation results on three generations of GPU machines demonstrate that the workload classification framework can tell apart the illicit authorized workloads with a high accuracy of over 95%.

Pengfei, Zou↗

Reinforcement Learning for Load-balanced Parallel Particle Tracing

We explore an online reinforcement learning (RL) paradigm to dynamically optimize parallel particle tracing performance in distributed-memory systems. Our method combines three novel components: (1) a work donation algorithm, (2) a high-order workload estimation model, and (3) a communication cost model. First, we design an RL-based work donation algorithm. Our algorithm monitors workloads of processes and creates RL agents to donate data blocks and particles from high-workload processes to low-workload processes to minimize program execution time. The agents learn the donation strategy on the fly based on reward and cost functions designed to consider processes' workload changes and data transfer costs of donation actions. Second, we propose a workload estimation model, helping RL agents estimate the workload distribution of processes in future computations. Third, we design a communication cost model that considers both block and particle data exchange costs, helping RL agents make effective decisions with minimized communication costs. We demonstrate that our algorithm adapts to different flow behaviors in large-scale fluid dynamics, ocean, and weather simulation data. Our algorithm improves parallel particle tracing performance in terms of parallel efficiency, load balance, and costs of I/O and communication for evaluations with up to 16,384 processors.

Distributed and parallel particle tracing↗

Benchmarking the Performance of Neuromorphic and Spiking Neural Network Simulators

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.

97 MATHEMATICS AND COMPUTING↗

Finding Your Niche: An Evolutionary Approach to HPC Topologies

Traditional interconnection network design approaches focus on building general network topologies by optimizing the bisection bandwidth or minimizing the network’s diameter to reduce the maximum distance between any two nodes, thus amortizing the overall execution time of the HPC workloads. While such network topologies may accommodate a wide variety of applications in general, this may result in sub-optimal performance for many frequently-executed or dynamic workloads. In this paper, instead of focusing on designing an all-encompassing, general-purpose network topology, we develop a methodology to design customized network interconnects, evolved by “finding” the optimal topologies for a particular target workload given by its communication and contention profiles. To this end, we implement a Genetic Algorithm (GA)-based approach for network topology design tailored to improve the overall execution time of a particular workload of interest. We conducted extensive experiments with well-known motifs in physics-based workloads (Sweep3D and FFT), as well as with a representative graph application (MiniVite), using the well-known Structural Simulation Toolkit (SST) Macroscale Element Library (SST/macro) simulator for network interconnect evaluation. We demonstrate that our genetic algorithm-based approach is robust enough to find the underlying optimal topology of a particular workload.

network interconnects, graph search, meta-heuristi↗

Automatic and Transparent Resource Contention Mitigation for Improving Large-Scale Parallel File System Performance

Proportional to the scale increases in HPC systems, many scientific applications are becoming increasingly data intensive, and parallel I/O has become one of the dominant factors impacting the large-scale HPC application performance. On a typical large-scale HPC system, we have observed that the lack of a global workload coordination coupled with the shared nature of storage systems cause load imbalance and resource contention over the end-to-end I/O paths resulting in severe performance degradation. I/O load imbalance on HPC systems is generally a self-inflicted wound and mostly occurs between the I/O paths and resources consumed by each individual job. In this paper, we introduce TAPP-IO, a dynamic, shared load balancing framework for mitigating resource contention. TAPP-IO extends our previous work and solves two major limitations: First, it transparently intercepts file creation calls during runtime to balance the workload over all available storage targets. The usage of TAPP-IO requires no application source code modifications and is independent from any I/O middleware. The framework can be applied to almost any HPC platform and is suitable for systems that lack a centralized file system resource manager. Second, the framework proposes a new placement strategy to support not only file-per-process I/O, but also single shared file I/O. This opens the door to a new class of scientific applications that can leverage the placement library for improved performance. We demonstrate the effectiveness of our integration on the Titan system at the Oak Ridge National Laboratory. Our experiments with a synthetic benchmark and real-world HPC workload show that, even in a noisy production environment, TAPP-IO can improve large-scale application performance significantly.

Neuwirth, Sarah↗

Frontier Job-Centric Telemetry Dataset

Comprehensive analysis of high-performance computing (HPC) systems requires linking workload execution to system behavior. This kind of analysis is vital for diagnosing performance issues, managing capacity, detecting anomalous workloads, and understanding how applications interact with system hardware. This job-centric telemetry dataset unifies scheduler job records with node-level measurements, enabling direct association between workloads and their corresponding power, thermal, and performance characteristics. It contains sanitized, scheduler related metadata for 152,400 individual jobs that ran on the Frontier supercomputer and ended on selected days throughout 2024 and 2025, a subpopulation of ~6.8% of the total number of allocated jobs with non-zero run time on the system over that same period. Each is linked with files that contain telemetry time series records of the power utilization and temperature behavior of its allocated nodes and their processors during the run time of the job. Where available, a portion of the job files also contain network performance time series. Jobs are sampled from select days that reflect normal levels of user activity and possess job size distributions with large numbers of leadership class jobs (>20% of Frontier nodes). Jobs in this dataset attempt to best represent successful user workflows.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

SMALE: Enhancing Scalability of Machine Learning Algorithms on Extreme-Scale Computing Platforms

Deployment and execution of machine learning tasks on extreme-scale computing platforms face several significant technical challenges: 1) High computing cost incurred by dense networks – The computing workload of deep networks with densely-connected topology increases rapidly with the network size, imposing a non-scalable computing model of extreme-scale computing platforms; 2) Non-optimized workload distribution – Many advanced deep learning algorithms, e.g., sparsification and irregular net-work topology, produce very unbalanced workload distribution on extreme-scale computing platforms. The computation efficiency is greatly hindered by the incurred data and computation redundancies as well as long tails of the node with extensive workload; 3) Constraints in data movement and I/O bottle-neck – Inter-node data movement in extreme-scale computing platforms are associated with high energy and latency costs, and subject to the constraints of I/O bandwidth; and 4) Generalization of algorithm realization and acceleration on computing platforms – The large varieties of machine learning algorithms and structures of extreme-scale computing platforms make the derivation of a generalized algorithm realization and acceleration method very challenging, which, however, is the requirement by domain scientists and interested users. We call the above challenges Smale’s Problems in Machine Learning and Understanding for High-Performance Computing Scientific Discovery. The objective of our three-year research project is to develop a holistic innovation set at structure, assembly, and acceleration layers of machine learning algorithms to address the above challenges in algorithm deployment and execution. Three tasks are particularly performed, including: At the algorithm structure level, we investigate the techniques that can structurally sparsify on the topology of deep networks for computing workload reduction. We also study clustering and pruning techniques that can optimize the workload distributions over the extreme-scale computing platforms; At the algorithm assembly level, we derive a unified learning framework for unsupervised transfer learning and dynamic growing capabilities. Novel training methods are also exploited to enhance the training efficiency of the proposed framework; At the algorithm acceleration level, we will develop a series of techniques that can accelerate the computation of sparse matrix operations, which are one of the core executions in deep learning and optimize memory access of the concerned platforms. Our proposed techniques attack the fundamental problems in machine learning algorithms running on extreme-scale computing platforms by vertically integrating the solutions at three closely entangled layers, paving the long-term scaling path of machine learning applications under DOE context. Three tasks corresponding to the above respective research orientations are performed during the three-year project period with our collaborators at ORNL. The outcome of the proposed project is anticipated to form a holistic solution set of novel algorithms and network topologies, efficient training techniques, and fast acceleration methods to promote the computing scalability of the machine learning applications of particular interest to DOE.

97 MATHEMATICS AND COMPUTING↗

Principles for problem aggregation and assignment in medium scale multiprocessors

One of the most important issues in parallel processing is the mapping of workload to processors. This paper considers a large class of problems having a high degree of potential fine grained parallelism, and execution requirements that are either not predictable, or are too costly to predict. The main issues in mapping such a problem onto medium scale multiprocessors are those of aggregation and assignment. We study a method of parameterized aggregation that makes few assumptions about the workload. The mapping of aggregate units of work onto processors is uniform, and exploits locality of workload intensity to balance the unknown workload. In general, a finer aggregate granularity leads to a better balance at the price of increased communication/synchronization costs; the aggregation parameters can be adjusted to find a reasonable granularity. The effectiveness of this scheme is demonstrated on three model problems: an adaptive one-dimensional fluid dynamics problem with message passing, a sparse triangular linear system solver on both a shared memory and a message-passing machine, and a two-dimensional time-driven battlefield simulation employing message passing. Using the model problems, the tradeoffs are studied between balanced workload and the communication/synchronization costs. Finally, an analytical model is used to explain why the method balances workload and minimizes the variance in system behavior.

Nicol, David M.↗

An analysis of scatter decomposition

A formal analysis of a powerful mapping technique known as scatter decomposition is presented. Scatter decomposition divides an irregular computational domain into a large number of equal sized pieces, and distributes them modularly among processors. A probabilistic model of workload in one dimension is used to formally explain why, and when scatter decomposition works. The first result is that if correlation in workload is a convex function of distance, then scattering a more finely decomposed domain yields a lower average processor workload variance. The second result shows that if the workload process is stationary Gaussian and the correlation function decreases linearly in distance until becoming zero and then remains zero, scattering a more finely decomposed domain yields a lower expected maximum processor workload. Finally it is shown that if the correlation function decreases linearly across the entire domain, then among all mappings that assign an equal number of domain pieces to each processor, scatter decomposition minimizes the average processor workload variance. The dependence of these results on the assumption of decreasing correlation is illustrated with situations where a coarser granularity actually achieves better load balance.

Nicol, David M.↗

Cockpit Adaptive Automation and Pilot Performance

The introduction of high-level automated systems in the aircraft cockpit has provided several benefits, e.g., new capabilities, enhanced operational efficiency, and reduced crew workload. At the same time, conventional 'static' automation has sometimes degraded human operator monitoring performance, increased workload, and reduced situation awareness. Adaptive automation represents an alternative to static automation. In this approach, task allocation between human operators and computer systems is flexible and context-dependent rather than static. Adaptive automation, or adaptive task allocation, is thought to provide for regulation of operator workload and performance, while preserving the benefits of static automation. In previous research we have reported beneficial effects of adaptive automation on the performance of both pilots and non-pilots of flight-related tasks. For adaptive systems to be viable, however, such benefits need to be examined jointly in the context of a single set of tasks. The studies carried out under this project evaluated a systematic method for combining different forms of adaptive automation. A model for effective combination of different forms of adaptive automation, based on matching adaptation to operator workload was proposed and tested. The model was evaluated in studies using IFR-rated pilots flying a general-aviation simulator. Performance, subjective, and physiological (heart rate variability, eye scan-paths) measures of workload were recorded. The studies compared workload-based adaptation to to non-adaptive control conditions and found evidence for systematic benefits of adaptive automation. The research provides an empirical basis for evaluating the effectiveness of adaptive automation in the cockpit. The results contribute to the development of design principles and guidelines for the implementation of adaptive automation in the cockpit, particularly in general aviation, and in other human-machine systems. Project goals were met or exceeded. The results of the research extended knowledge of automation-related performance decrements in pilots and demonstrated the positive effects of adaptive task allocation. In addition, several practical implications for cockpit automation design were drawn from the research conducted. A total of 12 articles deriving from the project were published.

Parasuraman, Raja↗

Effect of Dynamic Sector Boundary Changes on Air Traffic Controllers

The effect of dynamic sector boundary changes on air traffic controller workload was investigated with data from a human-in-the-loop simulation. Multiple boundary changes were made during simulated operations, and controller rating of workload was recorded. Analysis of these data showed an increase of 16.9% in controller workload due to boundary changes. This increased workload was correlated with the number of aircraft handoffs and change in sector volume. There was also a 12.7% increase in average workload due to the changed sector design after boundary changes. This increase was correlated to traffic flow crossing points getting closer to sector boundaries and an increase in the number of flights with short dwell time in a sector. This study has identified some of the factors that affect controller workload when sector boundaries are changed, but more research is needed to better understand their relationships.

Jung, Jaewoo↗

Novel Estimation of Pilot Performance Characteristics

Two mechanisms internal to the pilot that affect performance during a tracking task are: 1) Pilot equalization (i.e. lead/lag); and 2) Pilot gain (i.e. sensitivity to the error signal). For some applications McRuer's Crossover Model can be used to anticipate what equalization will be employed to control a vehicle's dynamics. McRuer also established approximate time delays associated with different types of equalization - the more cognitive processing that is required due to equalization difficulty, the larger the time delay. However, the Crossover Model does not predict what the pilot gain will be. A nonlinear pilot control technique, observed and coined by the authors as 'amplitude clipping', is shown to improve stability, performance, and reduce workload when employed with vehicle dynamics that require high lead compensation by the pilot. Combining linear and nonlinear methods a novel approach is used to measure the pilot control parameters when amplitude clipping is present, allowing precise measurement in real time of key pilot control parameters. Based on the results of an experiment which was designed to probe workload primary drivers, a method is developed that estimates pilot spare capacity from readily observable measures and is tested for generality using multi-axis flight data. This paper documents the initial steps to developing a novel, simple objective metric for assessing pilot workload and its variation over time across a wide variety of tasks. Additionally, it offers a tangible, easily implementable methodology for anticipating a pilot's operating parameters and workload, and an effective design tool. The model shows promise in being able to precisely predict the actual pilot settings and workload, and observed tolerance of pilot parameter variation over the course of operation. Finally, an approach is proposed for generating Cooper-Harper ratings based on the workload and parameter estimation methodology.

Human pilot models↗

Advanced Multimodal Solutions for Information Presentation

High-workload, fast-paced, and degraded sensory environments are the likeliest candidates to benefit from multimodal information presentation. For example, during EVA (Extra-Vehicular Activity) and telerobotic operations, the sensory restrictions associated with a space environment provide a major challenge to maintaining the situation awareness (SA) required for safe operations. Multimodal displays hold promise to enhance situation awareness and task performance by utilizing different sensory modalities and maximizing their effectiveness based on appropriate interaction between modalities. During EVA, the visual and auditory channels are likely to be the most utilized with tasks such as monitoring the visual environment, attending visual and auditory displays, and maintaining multichannel auditory communications. Previous studies have shown that compared to unimodal displays (spatial auditory or 2D visual), bimodal presentation of information can improve operator performance during simulated extravehicular activity on planetary surfaces for tasks as diverse as orientation, localization or docking, particularly when the visual environment is degraded or workload is increased. Tactile displays offer a third sensory channel that may both offload information processing effort and provide a means to capture attention when urgently required. For example, recent studies suggest that including tactile cues may result in increased orientation and alerting accuracy, improved task response time and decreased workload, as well as provide self-orientation cues in microgravity on the ISS (International Space Station). An important overall issue is that context-dependent factors like task complexity, sensory degradation, peripersonal vs. extrapersonal space operations, workload, experience level, and operator fatigue tend to vary greatly in complex real-world environments and it will be difficult to design a multimodal interface that performs well under all conditions. As a possible solution, adaptive systems have been proposed in which the information presented to the user changes as a function of taskcontext-dependent factors. However, this presupposes that adequate methods for detecting andor predicting such factors are developed. Further, research in adaptive systems for aviation suggests that they can sometimes serve to increase workload and reduce situational awareness. It will be critical to develop multimodal display guidelines that include consideration of smart systems that can select the best display method for a particular contextsituation.The scope of the current work is an analysis of potential multimodal display technologies for long duration missions and, in particular, will focus on their potential role in EVA activities. The review will address multimodal (combined visual, auditory andor tactile) displays investigated by NASA, industry, and DoD (Dept. of Defense). It also considers the need for adaptive information systems to accommodate a variety of operational contexts such as crew status (e.g., fatigue, workload level) and task environment (e.g., EVA, habitat, rover, spacecraft). Current approaches to guidelines and best practices for combining modalities for the most effective information displays are also reviewed. Potential issues in developing interface guidelines for the Exploration Information System (EIS) are briefly considered.

multimodal display↗

Exploring Human Factors Issues for Urban Air Mobility Operations

Urban air mobility (UAM) is currently receiving increased attention in the aviation literature as a new entrant into the airspace. Although the introduction of UAM offers the potential for significant benefits, it also creates the potential for fundamental change to the current air traffic management system. Several concepts are being explored to enable the development of a safe and efficient UAM system for near, mid and far term operations. A concept of operations for near term operations proposes several assumptions. Concepts for roles and responsibilities of human operators such as air traffic controllers propose different degrees of involvement. Identifying and exploring human factors issues is therefore a critical next step in the forward progression of concept development. A human-in-the-loop air traffic control simulation was used to investigate the effect of UAM traffic density and changes in current airspace routes and communication procedures on subjective controller workload and efficiency-related task performance. Findings indicate that although subjective workload was manageable for low density operations, medium and high density operations led to unmanageable levels of workload, leading to refusals to allow more vehicles into controlled airspace. By implementing a letter of agreement, verbal communications were reduced which were associated with reduced workload. Optimized routes were also associated with reduced workload and increased performance efficiency. Although these adjustments can positively support controller performance, workload still remained high during the high density UAM traffic scenarios. It is therefore suggested that, in order for UAM operation to become scalable, human operators will be required to work differently compared to current air traffic controllers. Future research should focus on the level and type of human operator or controller involvement and mated systems, to ensure safety and efficiency within UAM operations.

Air Traffic Management↗

Coordinating Garbage Collection for Arrays of Solid-state Drives

Although solid-state drives (SSDs) offer significant performance improvements over hard disk drives (HDDs) for a number of workloads, they can exhibit substantial variance in request latency and throughput as a result of garbage collection (GC). When GC conflicts with an I/O stream, the stream can make no forward progress until the GC cycle completes. GC cycles are scheduled by logic internal to the SSD based on several factors such as the pattern, frequency, and volume of write requests. When SSDs are used in a RAID with currently available technology, the lack of coordination of the SSD-local GC cycles amplifies this performance variance. We propose a global garbage collection (GGC) mechanism to improve response times and reduce performance variability for a RAID of SSDs. We include a high-level design of SSD-aware RAID controller and GGC-capable SSD devices and algorithms to coordinate the GGC cycles. We develop reactive and proactive GC coordination algorithms and evaluate their I/O performance and block erase counts for various workloads. Our simulations show that GC coordination by a reactive scheme improves average response time and reduces performance variability for a wide variety of enterprise workloads. For bursty, write-dominated workloads, response time was improved by 69% and performance variability was reduced by 71%. We show that a proactive GC coordination algorithm can further improve the I/O response times by up to 9% and the performance variability by up to 15%. We also observe that it could increase the lifetimes of SSDs with some workloads (e.g. Financial) by reducing the number of block erase counts by up to 79% relative to a reactive algorithm for write-dominant enterprise workloads.

Flash Memory↗

Spatial Locality-Aware Cache Partitioning for Effective Cache Sharing, In: 2015 44th International Conference on Parallel Processing

In modern multi-core processors, last-level caches (LLCs) are typically shared among multiple cores. Previous works have shown that such sharing is beneficial as different workloads have different needs for cache capacity, and logical partitioning of capacity can improve system performance. However, what is missing in previous works on partitioning shared LLCs is that the heterogeneity in spatial locality among workloads has not been explored. In other words, all the cores use the same block/line size in shared LLCs. In this work, we highlight that exploiting spatial locality enables much more effective cache sharing. The fundamental reason is that for many memory intensive workloads, their cache capacity requirements can be drastically reduced when a large block size is employed, therefore they can effectively donate more capacity to other workloads. To leverage spatial locality for cache partitioning effectively, we first propose a simple yet effective mechanism to measure both spatial and temporal locality at run-time. The locality information is then used to determine both the proper block size and the capacity assigned to each workload. Our experiments show that our Spatial Locality-aware Cache Partitioning (SLCP) significantly outperforms the previous works. We also present several case studies that dissect the effectiveness of SLCP compared to the existing approaches.

Computer Science↗

MAPA: Multi-Accelerator Pattern Allocation Policy for Multi-Tenant GPU Servers

Multi-accelerator servers are increasingly being deployed in shared multi-tenant environments (such as in cloud data centers) in order to meet the demands of large-scale compute-intensive workloads. In addition, these accelerators are increasingly being inter-connected in complex topologies and workloads are exhibiting a wider variety of inter-accelerator communication patterns. However, existing al-location policies are ill-suited for these emerging use-cases. Specifically, this work identifies that multi-accelerator workloads are commonly fragmented leading to reduced bandwidth and increased latency for inter-accelerator communication. We propose Multi-Accelerator Pattern Allocation (MAPA), a graph pattern mining approach towards providing generalized allocation support for allocating multi-accelerator workloads on multi-accelerator servers. We demonstrate that MAPA is able to improve the execution time of multi-accelerator workloads and that MAPA is able to provide generalized benefits across various accelerator topologies. Finally, we demonstrate a speedup of 12.4% for75th percentile of jobs with the worst case execution time reduced by up to 35% against baseline policy using MAPA.

Ranganath, Kiran↗

AI4IO: A suite of AI-based tools for IO-aware scheduling

Traditional workload managers do not have the capacity to consider how IO contention can increase job runtime and even cause entire resource allocations to be wasted. Whether from bursts of IO demand or parallel file systems (PFS) performance degradation, IO contention must be identified and addressed to ensure maximum performance. In this paper, we present AI4IO (AI for IO), a suite of tools using AI methods to prevent and mitigate performance losses due to IO contention. AI4IO enables existing workload managers to become IO-aware. Currently, AI4IO consists of two tools: PRIONN and CanarIO. PRIONN predicts IO contention and empowers schedulers to prevent it. CanarIO mitigates the impact of IO contention when it does occur. We measure the effectiveness of AI4IO when integrated into Flux, a next-generation scheduler, for both small- and large-scale IO-intensive job workloads. Our results show that integrating AI4IO into Flux improves the workload makespan up to 6.4%, which can account for more than 18,000 node-h of saved resources per week on a production cluster in our large-scale workload.

Wyatt, II, Michael R.↗