Search NASA⌕ Search

SEARCH · Search NASA

Results for “Resource scheduling”

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 505 records · Page 28

Simple Linux Utility for Resource Management

SLURM is an open source, fault-tolerant, and highly scalable cluster management and job scheduling system for large and small computer clusters. As a cluster resource manager, SLURM has three key functions. First, it allocates exclusive and/or non exclusive access to resources (compute nodes) to users for some duration of time so they can perform work. Second, it provides a framework for starting, executing, and monitoring work (normally a parallel job) on the set of allciated nodes. Finally, it arbitrates conflicting requests for resouces by managing a queue of pending work.

Jette, M.↗

Batch Scheduling a Fresh Approach

The Network Queueing System (NQS) was designed to schedule jobs based on limits within queues. As systems obtain more memory, the number of queues increased to take advantage of the added memory resource. The problem now becomes too many queues. Having a large number of queues provides users with the capability to gain an unfair advantage over other users by tailoring their job to fit in an empty queue. Additionally, the large number of queues becomes confusing to the user community. The High Speed Processors group at the Numerical Aerodynamics Simulation (NAS) Facility at NASA Ames Research Center developed a new approach to batch job scheduling. This new method reduces the number of queues required by eliminating the need for queues based on resource limits. The scheduler examines each request for necessary resources before initiating the job. Also additional user limits at the complex level were added to provide a fairness to all users. Additional tools which include user job reordering are under development to work with the new scheduler. This paper discusses the objectives, design and implementation results of this new scheduler

Cardo, Nicholas P.↗

Facility Utilization and Risk Analysis for Remediation of Legacy Transuranic Waste at the Savannah River Site - 13572

Savannah River Nuclear Solutions (SRNS) completed the Accelerated TRU Project for remediating legacy waste at the Savannah River Site with significant cost and schedule efficiencies due to early identification of resources and utilization of risk matrices. Initial project planning included identification of existing facilities that could be modified to meet the technical requirements needed for repackaging and remediating the waste. The project schedule was then optimized by utilization of risk matrices that identified alternate strategies and parallel processing paths which drove the overall success of the project. Early completion of the Accelerated TRU Project allowed SRNS to pursue stretch goals associated with remediating very difficult TRU waste such as concrete casks from the hot cells in the Savannah River National Laboratory. Project planning for stretch goals also utilized existing facilities and the risk matrices. The Accelerated TRU project and stretch goals were funded under the American Recovery and Reinvestment Act (ARRA). (authors)

54 ENVIRONMENTAL SCIENCES↗

End-to-end planning and scheduling systems technology for space operations

Consideration is given to planning and scheduling operations concepts from an end-to-end perspective, through both mission operations and institutional support functions. An operations concept is proposed which is based on a flexible request language used to state resource requirements and mission constraints to a scheduling system. The language has the potential to evolve into an international standard for exchanging service request information on international space networks. The key benefit of the flexible scheduling request concept is the shift of a significant conflict resolution effort from humans to computers, reducing the time for generating a week's worth of schedules to hours instead of days.

Moe, Karen L.↗

A Secure and Adaptive Hierarchical Multi-Timescale Framework for Resilient Load Restoration Using a Community Microgrid

Distribution system integrated community microgrids (CMGs) can partake in restoring loads during extended duration outages. At such times, the CMGs are challenged with limited resource availability, absence of robust grid support, and heightened demand-supply uncertainty. Here, this paper proposes a secure and adaptive three-stage hierarchical multi-timescale framework for scheduling and real-time (RT) dispatch of CMGs with hybrid PV systems to address these challenges. The framework enables the CMG to dynamically expand its boundary to support the neighboring grid sections and is adaptive to the changing forecast error impacts. The first stage solves a stochastic extended duration scheduling (EDS) problem to obtain referral plans for optimal resource rationing. The intermediate near-real-time (NRT) scheduling stage updates the EDS schedule closer to the dispatch time using new obtained forecasts, followed by the RT dispatch stage. To make the decisions more secure and robust against forecast errors, a novel concept called delayed recourse is designed. The approach is evaluated via numerical simulations on a modified IEEE 123-bus system and validated using OpenDSS and hardware-in-loop simulations. The results show superior performance in maximizing load supply and continuous secure distribution network operation under different operating scenarios.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Link Scheduling in Satellite Networks via Machine Learning Over Riemannian Manifolds

Low Earth Orbit (LEO) satellites play a crucial role in enhancing global connectivity, serving a complementary solution to existing terrestrial systems. In wireless networks, scheduling is a vital process that allocates time-frequency resources to users for interference management. However, LEO satellite networks face significant challenges in scheduling their links towards ground users due to the satellites’ mobility and overlapping coverage. This paper addresses the dynamic link scheduling problem in LEO satellite networks by considering spatio-temporal correlations introduced by the satellites’ movements. The first step in the proposed solution involves modeling the network over Riemannian manifolds, thanks to their representation as symmetric positive definite matrices. We introduce two machine learning (ML)-based link scheduling techniques that model the dynamic evolution of satellite positions and link conditions over time and space. To accurately predict satellite link states, we present a recurrent neural network (RNN) over Riemannian manifolds, which captures spatio-temporal characteristics over time. Furthermore, we introduce a separate model, the convolutional neural network (CNN) over Riemannian manifolds, which captures geometric relationships between satellites and users by extracting spatial features from the network topology across all links. Simulation results demonstrate that both RNN and CNN over Riemannian manifolds deliver comparable performance to the fractional programming-based link scheduling (FPLinQ) benchmark. Remarkably, unlike other ML-based models that require extensive training data, both models only need 30 training samples to achieve over 99% of the sum rate while maintaining similar computational complexity relative to the benchmark.

42 ENGINEERING↗

Predicting Computer System Failures Using Support Vector Machines

Mitigating the impact of computer failure is possible if accurate failure predictions are provided. Resources, applications, and services can be scheduled around predicted failure and limit the impact. Such strategies are especially important for multi-computer systems, such as compute clusters, that experience a higher rate failure due to the large number of components. However providing accurate predictions with sufficient lead time remains a challenging problem. This paper describes a new spectrum-kernel Support Vector Machine (SVM) approach to predict failure events based on system log files. These files contain messages that represent a change of system state. While a single message in the file may not be sufficient for predicting failure, a sequence or pattern of messages may be. The approach described in this paper will use a sliding window (sub-sequence) of messages to predict the likelihood of failure. The frequency representation of the message sub-sequences observed are then used as input to the SVM that associates the messages to a class of failed or non-failed system. Experimental results using actual system log files from a Linux-based compute cluster indicate the proposed SVM approach can predict hard disk failure with an accuracy of 76% one day in advance.

99 GENERAL AND MISCELLANEOUS↗

Concepts in Distributed Scheduling and Control

To support instrument and experiment operations effectively in the Space Station era, planning, scheduling and control must allow for: interactive, realtime, remote operations; responsive scheduling and rescheduling; support of the full range of distributed science, application and commercial users; interaction and cooperation among distributed users; and efficient use of on-board, communications, and ground-based resources. We suggest conceptual and managerial approaches that address these needs. Specifically, we describe an approach to distributed planning, scheduling and control functions that is based on resources and on a distributed knowledge hierarchy. We describe these functions as components of an integrated management system. We discuss automated scheduling assistants and integration of planning and scheduling functions with realtime operations control. The suggested approach, taken from a users' point-of-view, has resulted in the Science User Resource Planning and Scheduling System (SURPASS). In this paper, we describe the major components of SURPASS and discuss the features of this innovative prototype. Further ideas concerning instrument planning, scheduling, and control may be found in the Space Station Instrument Control System Study.

Hansen, Elaine R.↗

A hybrid architecture for volt-var control in active distribution grids

Modern active distribution grids are characterized by the increasing penetration of distributed energy resources (DERs). The proper coordination and scheduling of a large numbers of these small-scale and spatially distributed DERs is necessary, and warrants the use of novel distributed approaches. In this paper, we propose a hybrid volt-var control architecture for the distribution grid, which leverages existing centralized and local approaches to planning, decision making, and control, and augments it with distributed optimization and distributed control for DER management. First, we propose a convex model to describe the power physics of distribution grids of meshed topology and unbalanced structure, based on current injection and McCormick Envelopes. Second, we employ the distributed proximal atomic coordination (PAC) algorithm to coordinate DERs to provide voltage support. We implement volt-var optimization by optimally coordinating DERs including PV smart inverters and demand response. We present results using the IEEE-34 bus network, using real data from a distribution feeder in Hawaii, to model load and PV generation. Different levels of DER penetration and objective functions are simulated. Finally, our results show the need for the coordination of DERs to improve voltage profiles, even in networks with existing voltage control devices. Further, we show the need for flexible reactive power capabilities to achieve desired grid performance.

24 POWER TRANSMISSION AND DISTRIBUTION↗

The ePIC Simulation Campaign Workflow on the Open Science Grid

The ePIC collaboration is realizing the first experiment of the future Electron-Ion Collider (EIC) at the Brookhaven National Laboratory that will allow for a precision study of the nucleons and the nucleus at the scale of sea quarks and gluons through the study of electron-proton/ion collisions. This paper will discuss the current workflow for running centralized simulation campaigns for ePIC on the Open Science Grid (OSG) infrastructure. This involves monthly releases of ePIC software and container deployments to CVMFS, generation of input datasets in HepMC format according to collaboration-defined policy, using Snakemake in CI/CD for validation and benchmarking, and submitting jobs to the OSG condor scheduler for opportunistic running on available resources. File transfers utilize XrootD, and Rucio is used for data management. The workflow is continuously refined to improve daily throughput (currently 50-100k core hours per day) and minimize job failures. Since May 2023, monthly simulation campaigns employing the workflow have cumulatively used over 20 million core hours on the OSG and produced over 350 TB of simulation data. The campaigns incorporate simulations for the broad science program of the EIC and are actively used for the detector and physics studies in preparation of the Technical Design Report (TDR).

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Induced Radioactivity in Lead Shielding at the National Synchrotron Light Source

The National Synchrotron Light Source (NSLS) at Brookhaven National Laboratory was shut down in September 2014. Lead bricks used as radiological shadow shielding within the accelerator were exposed to stray radiation fields during normal operations. The FLUKA code, a fully integrated Monte Carlo simulation package for the interaction and transport of particles and nuclei in matter, was used to estimate induced radioactivity in this shielding and stainless steel beam pipe from known beam losses. The FLUKA output was processed using MICROSHIELD® to estimate on-contact exposure rates with individually exposed bricks to help design and optimize the radiological survey process. This entire process can be modeled using FLUKA, but use of MICROSHIELD® as a secondary method was chosen because of the project’s resource constraints. Due to the compressed schedule and lack of shielding configuration data, simple FLUKA models were developed in this paper. FLUKA activity estimates for stainless steel were compared with sampling data to validate results, which show that simple FLUKA models and irradiation geometries can be used to predict radioactivity inventories accurately in exposed materials. During decommissioning 0.1% of the lead bricks were found to have measurable levels of induced radioactivity. Finally, post-processing with MICROSHIELD® provides an acceptable secondary method of estimating residual exposure rates.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗

Elastic distributed training with fast convergence and efficient resource utilization

Distributed learning is now routinely conducted on cloud as well as dedicated clusters. Training with elastic resources brings new challenges and design choices. Prior studies focus on runtime performance and assume a static algorithmic behavior. In this work, by analyzing the impact of of resource scaling on convergence, we introduce schedules for synchronous stochastic gradient descent that proactively adapt the number of learners to reduce training time and improve convergence. Our approach no longer assumes a constant number of processors throughout training. In our experiment, distributed stochastic gradient descent with dynamic schedules and reduction momentum achieves better convergence and significant speedups over prior static ones. Numerous distributed training jobs running on cloud may benefit from our approach.

Cong, Guojing↗

Stochastic Optimization for Unit Commitment-A Review

Optimization models have been widely used in the power industry to aid the decision-making process of scheduling and dispatching electric power generation resources, a process known as unit commitment (UC). Since UC's birth, there have been two major waves of revolution on UC research and real life practice. The first wave has made mixed integer programming stand out from the early solution and modeling approaches for deterministic UC, such as priority list, dynamic programming, and Lagrangian relaxation. With the high penetration of renewable energy, increasing deregulation of the electricity industry, and growing demands on system reliability, the next wave is focused on transitioning from traditional deterministic approaches to stochastic optimization for unit commitment. Since the literature has grown rapidly in the past several years, this paper is to review the works that have contributed to the modeling and computational aspects of stochastic optimization (SO) based UC. Relevant lines of future research are also discussed to help transform research advances into real-world applications.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Model Formulations: Market clearing models, market design specifications, and dispatch simulation

This document provides various model formulations that will be implemented for the Energy Storage Participation Algorithm pilot competition (ESPA-Comp). This competition will assess the performance of different storage offer algorithms in terms of their ability to maximize the value of storage resources participating under various market designs that vary in the range of complexity of possible storage offers. The model formulations include a general market clearing optimization model and specification to implement it in various market designs. We provide specifications for three market designs to be tested in ESPA-Comp, which we call the two-settlement, multi-settlement, and rolling horizon forward markets. The market designs include different trading frequencies and offer formats. Storage resources participating in the market are provided with primitive values for the resource’s physical capabilities, and they are provided with a storage offer format that their algorithms will be tasked with populating. A detailed physical model is used to assess each storage resource’s ability to maintain its scheduled dispatch according to its state-of-charge, operating temperature, and power conversion efficiency.

25 ENERGY STORAGE↗

Space station human productivity study. Volume 5: Management plans

The 67 Management Plans represent recommended study approaches for resolving 108 of the 305 Issues which were identified. Each study Management Plan is prepared in three formats: Management Plan Overview (lists the subsumed Issues, study background, and related overview information); Study Plan (details the study approach by tasks, lists special needs, and describes expected study products); Schedule-Task Flow (provides a time-lined schedule for the study tasks and resource requirements). The Management Relationships Matrix, included in this volume, shows the data input-output relationships among all recommended studies. A listing is also included which cross-references the unresolved requirements to Issues to management plans. A glossary of all abbreviations utilized is provided.

Source record↗

Software design with fuzzy requirements (A case study)

The author describes the resource allocation and planning helper (RALPH) scheduling system developed at NASA's Jet Propulsion Laboratory. RALPH addresses the concerns of designing software systems to minimize the need to recode for changes and upgrades; this concern is acute when requirements are uncertain or changing. Determining requirements, understanding the problem, designing for change, and tradeoffs are also discussed.

Werntz, David G.↗