Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Jobs”

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 253 records · Page 14
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Cost-Benefit Analysis For Indonesia Building Sector: Whole-Building Cooling Solutions

The Net Zero World (NZW) Initiative Collaborative Work Program with the Government of Indonesia (GoI) includes technical assistance and investment mobilization facilitation to accelerate deployment of energy efficiency technologies and solutions for the building sector. A February 2023 U.S.–Indonesia Joint Workshop on Decarbonizing the Building Sector yielded a NZW Indonesia Building Decarbonization Working Group (NZW IBDWG) with four sub-working groups (SWG): SWG-A National Center, SWG-B Capacity Building, SWG-C Investment and Financing, and SWG-D Pilot Projects. Technical analysis of whole-building cooling solutions for tropical climates of Indonesia was conducted by SWG-A to quantify energy savings, carbon dioxide reductions, and comfort improvements offered by 12 passive or low-energy cooling strategies: ceiling fans with and without thermostat setbacks; cool roofs; cool walls; exterior awnings; exterior shades; interior shades; insulated roofs; insulated walls; low-e windows; solar window films; and natural ventilation. Leveraging the results from SWG-A, cost-benefit analysis (CBA) was conducted by SWG-C to assess the consumer and national costs and impacts associated with these 12 cooling solutions. The evaluation involved estimating life-cycle costs (LCC), payback period (PBP), net present values (NPV), annual electricity burden change for low-income households, and reduced national annual power-sector generation demand by 2030, 2040, 2050, and 2060. This evaluation can help guide Indonesia’s Just Energy Transition Partnership (JETP) investments in policies and programs to advance research, development, deployment, and commercial adoption (RDDCA) of efficient residential building sector cooling technologies and solutions in Indonesia. Four key energy conservation measures (ECM) have been identified to reduce air-conditioning (AC) energy demand in single-family housing in Indonesia: ceiling fan with temperature setback (to 28.1 °Celcius from 25 °C); insulated walls; insulated roof; and cool roof. This study found that low-income households with AC installations in Indonesia currently face a high energy cost burden of approximately 10%. However, by implementing a ceiling fan with temperature setback, this burden could decrease to 2.5% today and further reduce to 1.3% by the year 2060. The PBP for a ceiling fan with temperature setback is one year, indicating one of the lowest LCC and best NPV. In the planned upcoming phase of CBA, a series of building cooling improvement scenarios can be further defined, incorporating more than one ECM in combination with socio-economic factors evaluated in the initial CBA phase. Additionally, the analysis of ECM effects in multifamily housing can be expanded. This broader national analysis aims to encompass a holistic and comprehensive system-level perspective, including factors such as avoided power sector infrastructure investments, domestic job creation, domestic manufacturing job creation, and gross domestic product (GDP) growth.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2024

A Mid-Century Net-Zero Scenario for the State of Wyoming and its Economic Impacts

Clean hydrogen has the potential to help achieve 10% economy-wide emissions reductions by 2050 relative to 2005, promote energy security and resilience, and develop a new economy in the United States. In 2030, the hydrogen economy could create about 100,000 new jobs to build new capital projects and clean hydrogen infrastructure. The Wyoming Energy Authority recently announced the state’s energy strategy, which establishes a goal of net-zero emissions by 2050. Under all likely scenarios, achieving a mid-century net-zero target will pose challenges and create opportunities for Wyoming’s energy sector. If executed properly, the transition could favorably affect the state’s economy overall in the long term. This research program examines the economic impact of fossil energy production in Wyoming and provides various predictions for future energy mixes to achieve net-zero emissions. Preliminary work suggests that Wyoming-based hydrogen production could have significant economic benefits and job creation implications for Wyoming. This study further assesses Wyoming’s opportunities to create hydrogen-based industries, assess economic impacts, identify knowledge gaps and research needs, and create a Hydrogen Center of Excellence to accelerate commercialization and deployment. This project helped to understand Wyoming's areas of focus for research and development and identified its areas of strength and potential challenges in creating a hydrogen ecosystem. As a result of this study, we estimate that for blue hydrogen produced from coal and gas resources, the overall cost reduction will be driven mainly by the carbon-sequestration tax credit and the improvement in carbon capture. Mature technologies, like SMR and PSA, will make limited contributions. They have no or limited reductions from an additional capacity deployment in future costs. We also understand the importance of continued support from public and private sectors for Carbon Capture and Storage (CCS)-related research, development, and demonstration programs at federal and state levels. The successful and efficient production of blue hydrogen requires a unique blend of energy resources, geology, regulation, law, and infrastructure. Wyoming has the distinction of meeting all these demands. The team also estimates that the availability and command of water resources accessible for hydrogen production are crucial for developing new projects. Water treatment, use, and disposal after treatment will also make projects possible. Primarily, this is relevant for hydrogen made using renewable energy. Wyoming has one of the best wind resource capacity in the nation. Harnessing this resource is challenging due to limited transmission line availability. Hydrogen could become one of the solutions to the stranded resource problem, primarily if the water availability challenge is addressed. Using produced oil & gas water could help to solve the problem. A commonly cited barrier to the expansion of hydrogen markets is the cost associated with constructing new pipelines, which typically require large amounts of capital to develop. Wyoming already possesses much of the export infrastructure needed to connect Wyoming’s hydrogen production with major markets across the West Coast, Pacific Northwest, Midwest, and Front Range regions of the United States, where a large portion of Wyoming’s natural gas is already transported. In addition to transportation by pipeline, rail transportation of hydrogen has also proven feasible. Wyoming uses its extensive railway system to transport large amounts of coal to its export partners across the United States. By using cryogenic or compressed-gas cars, Wyoming has the potential to add hydrogen to its existing network of railroad energy exports. The same technology may also be applied to hydrogen transport via trucks traveling interstate highways. Wyoming’s workforce is ready to meet the demands of clean hydrogen development. Many of the skills and training needed for hydrogen production are the same skills already possessed by Wyoming’s oil & gas and coal workforce. Many government and industry leaders expect clean hydrogen and other low-carbon energy projects to generate significant job growth and to recruit many already-trained oil & gas and coal workers whose jobs may be displaced. As energy companies seek to penetrate the markets for Wyoming hydrogen production, there is a natural mutual benefit to Wyoming’s workers and companies seeking to launch projects with the assistance of a trained workforce. Wyoming’s university and community college system have adopted several programs to ensure that highly qualified engineers and other technically skilled employees continue to graduate with skills to support the development of hydrogen and other innovative energy projects moving forward. Throughout the project, stakeholder outreach and education took many forms, including meetings with several major companies in the industry, collaborating with local government organizations, educational organizations, and national laboratories, tribal outreach and engagement, the sponsoring of several hydrogen-focused projects in many departments throughout the University of Wyoming, and developing a collaboration with international universities. The products of these collaborations consist of working relationships with several companies in the industry, educational institutions, national labs, and local government, as well as strong connections with individuals who will play an essential role in the success of the Hydrogen Energy Research Center.

08 HYDROGEN↗
DOE OSTI2024

A Managed Tokens Service for Securely Keeping and Distributing Grid Tokens

Fermilab is transitioning authentication and authorization for grid operations to using bearer tokens based on the WLCG Common JWT (JSON Web Token) Profile. One of the functionalities that Fermilab experimenters rely on is the ability to automate batch job submission, which in turn depends on the ability to securely refresh and distribute the necessary credentials to experiment job submit points. Thus, with the transition to using tokens for grid operations, we needed to create a service that would obtain, refresh, and distribute tokens for experimenters’ use. This service would avoid the need for experimenters to be experts in obtaining their own tokens and would better protect the most sensitive long-lived credentials. Further, the service needed to be widely scalable, as we are currently keeping credentials active for approximately 15 experiments, each with 1-3 different credentials, and distributing those credentials to 2-20 submit points per experiment, with those numbers steadily increasing. To address these issues, we created and deployed a Managed Tokens service. The service is written in Go, taking advantage of that language’s native concurrency primitives to easily be able to scale operations as we onboard experiments. The service uses as its first credentials a set of kerberos keytabs, stored on the same secure machine that the Managed Tokens service runs on. These kerberos credentials allow the service to use htgettoken via condor_vault_storer to store vault tokens in the HTCondor credential managers (credds) that run on the batch system scheduler machines (HTCondor schedds); as well as downloading a local, shorter-lived copy of the vault token. The kerberos credentials are then also used to distribute copies of the locally-stored vault tokens to experiment submit points. When experimenters schedule jobs to be submitted, these distributed vault tokens are used to access a Hashicorp Vault instance (run separately from the Managed Tokens service), and previously-stored refresh tokens there are used to obtain the bearer token that is submitted with the job. We will discuss here the design of the Managed Tokens service, including elaborating on certain choices we made with regards to concurrent operations, configuration, monitoring, and deployment.

Bhat, Shreyas↗
DOE OSTI2025

Decentralized Distributed Proximal Policy Optimization (DD-PPO) for High Performance Computing Scheduling on Multi-User Systems

Resource allocation in High Performance Computing (HPC) environments presents a complex and multifaceted challenge for job scheduling algorithms. Beyond the efficient allocation of system resources, schedulers must account for and optimize multiple performance metrics, including job wait time and system throughput. Traditional heuristic-based scheduling algorithms increasingly struggle and lack the efficiency needed to meet the demands and address the complexity and scale of modern HPC systems. Consequently, recent research efforts have focused on leveraging advancements in Artificial Intelligence (AI) and Deep Learning (DL), particularly Reinforcement Learning (RL), to develop more adaptable and intelligent scheduling strategies. Previous RL-based scheduling approaches have explored a range of algorithms, from Deep Q-Networks (DQN) to Proximal Policy Optimization (PPO), and more recently, hybrid methods that integrate Graph Neural Networks (GNNs) with RL techniques. However, a common limitation across these methods is their reliance on relatively small datasets, with few methods being evaluated using large-scale, multi-million-job trace datasets representative of real-world HPC workloads. Moreover, existing RL schedulers face scalability issues due to centralized policy updates, which hinder training efficiency and performance when applied to large datasets. This study introduces a novel RL-based scheduler utilizing Decentralized Distributed Proximal Policy Optimization (DD-PPO) algorithm, which supports large-scale distributed training across multiple workers without requiring parameter synchronization at every step. By eliminating reliance on centralized updates to a shared policy, the DD-PPO scheduler enhances scalability, training efficiency, and sample utilization. Experimental validation using a large real-world dataset containing over 11.5 million job traces collected from petascale HPC systems over six years assesses the influence of dataset scale on training effectiveness and compares DD-PPO performance to traditional and advanced scheduling approaches. The experimental results demonstrate improved scheduling performance in comparison to both heuristic-based schedulers and existing RL-based scheduling algorithms.

AI↗
NASA NTRS1972

What motivates researchers in times of economic uncertainty.

Results of a study initiated late in 1970 to obtain both a measure of on-and-around-the-job factors which were 'motivating' to engineers and scientists, and to obtain an indication of how the relative importance of these factors changes as a result of the uncertain economic environment. A questionnaire, 'The Jackman Job Satisfaction Schedule,' was used to satisfy the needs of the study. It is concluded that managers can enhance the feeling of motivation by making individual job assignments interesting and challenging, by formulating significant milestones and end points into job content, and by assigning ample rewards with corresponding responsibility. In times of economic uncertainty increased emphasis should be given to security-related aspects of employment.

Bucher, G. C.↗
NASA NTRS1969

Analysis of a Multiprocessor Guidance Computer

The design of the next generation of spaceborne digital computers is described. It analyzes a possible multiprocessor computer configuration. For the analysis, a set of representative space computing tasks was abstracted from the Lunar Module Guidance Computer programs as executed during the lunar landing, from the Apollo program. This computer performs at this time about 24 concurrent functions, with iteration rates from 10 times per second to once every two seconds. These jobs were tabulated in a machine-independent form, and statistics of the overall job set were obtained. It was concluded, based on a comparison of simulation and Markov results, that the Markov process analysis is accurate in predicting overall trends and in configuration comparisons, but does not provide useful detailed information in specific situations. Using both types of analysis, it was determined that the job scheduling function is a critical one for efficiency of the multiprocessor. It is recommended that research into the area of automatic job scheduling be performed.

Maltach, E. G.↗
NASA NTRS1976

Simulating an aerospace multiprocessor

The paper describes a simulator which was used to evaluate the architecture of an aerospace multiprocessor. The simulator models interactions among the processors, memories, the central data bus, and a possible 'job stack'. Special features of the simulator are discussed, including the use of explicitly coded and individually distinguishable 'job models' instead of a statistically defined 'job mix' and a specialized Job Model Definition Language to automate the detailed coding of the models. Some results are presented which show that when the simulator was employed in conjunction with queuing theory and Markov-process analysis, more insight into system behavior was obtained than would have been with any one technique alone.

Mallach, E. G.↗
NASA NTRS1980

The Remote Computer Control (RCC) system

A system to remotely control job flow on a host computer from any touchtone telephone is briefly described. Using this system a computer programmer can submit jobs to a host computer from any touchtone telephone. In addition the system can be instructed by the user to call back when a job is finished. Because of this system every touchtone telephone becomes a conversant computer peripheral. This system known as the Remote Computer Control (RCC) system utilizes touchtone input, touchtone output, voice input, and voice output. The RCC system is microprocessor based and is currently using the INTEL 80/30microcomputer. Using the RCC system a user can submit, cancel, and check the status of jobs on a host computer. The RCC system peripherals consist of a CRT for operator control, a printer for logging all activity, mass storage for the storage of user parameters, and a PROM card for program storage.

Holmes, W.↗
NASA NTRS1987

IBM JCL made easy by means of ISPF dialog application

An interactive cathode ray tube (CRT) terminal application has been developed which automatically generates the job control language (JCL) and submits NASTRAN batch jobs at an IBM mainframe installation. The relevant parameters for each NASTRAN batch job are stored in a table. Once a table entry is defined, the corresponding NASTRAN batch job may be submitted as often as desired with a minimum of keystrokes. For most users, there is no need for any knowledge of JCL.

Cooper, Charles E.↗
NASA NTRS1991

An expert system for planning and scheduling in a telerobotic environment

A knowledge based approach to assigning tasks to multi-agents working cooperatively in jobs that require a telerobot in the loop was developed. The generality of the approach allows for such a concept to be applied in a nonteleoperational domain. The planning architecture known as the task oriented planner (TOP) uses the principle of flow mechanism and the concept of planning by deliberation to preserve and use knowledge about a particular task. The TOP is an open ended architecture developed with a NEXPERT expert system shell and its knowledge organization allows for indirect consultation at various levels of task abstraction. Considering that a telerobot operates in a hostile and nonstructured environment, task scheduling should respond to environmental changes. A general heuristic was developed for scheduling jobs with the TOP system. The technique is not to optimize a given scheduling criterion as in classical job and/or flow shop problems. For a teleoperation job schedule, criteria are situation dependent. A criterion selection is fuzzily embedded in the task-skill matrix computation. However, goal achievement with minimum expected risk to the human operator is emphasized.

Ntuen, Celestine A.↗
NASA NTRS1992

Generating effective project scheduling heuristics by abstraction and reconstitution

A project scheduling problem consists of a finite set of jobs, each with fixed integer duration, requiring one or more resources such as personnel or equipment, and each subject to a set of precedence relations, which specify allowable job orderings, and a set of mutual exclusion relations, which specify jobs that cannot overlap. No job can be interrupted once started. The objective is to minimize project duration. This objective arises in nearly every large construction project--from software to hardware to buildings. Because such project scheduling problems are NP-hard, they are typically solved by branch-and-bound algorithms. In these algorithms, lower-bound duration estimates (admissible heuristics) are used to improve efficiency. One way to obtain an admissible heuristic is to remove (abstract) all resources and mutual exclusion constraints and then obtain the minimal project duration for the abstracted problem; this minimal duration is the admissible heuristic. Although such abstracted problems can be solved efficiently, they yield inaccurate admissible heuristics precisely because those constraints that are central to solving the original problem are abstracted. This paper describes a method to reconstitute the abstracted constraints back into the solution to the abstracted problem while maintaining efficiency, thereby generating better admissible heuristics. Our results suggest that reconstitution can make good admissible heuristics even better.

Janakiraman, Bhaskar↗
NASA NTRS1996

Trends and Issues in Fuzzy Control and Neuro-Fuzzy Modeling

Everyday experience in building and repairing things around the home have taught us the importance of using the right tool for the right job. Although we tend to think of a 'job' in broad terms, such as 'build a bookcase,' we understand well that the 'right job' associated with each 'right tool' is typically a narrowly bounded subtask, such as 'tighten the screws.' Unfortunately, we often lose sight of this principle when solving engineering problems; we treat a broadly defined problem, such as controlling or modeling a system, as a narrow one that has a single 'right tool' (e.g., linear analysis, fuzzy logic, neural network). We need to recognize that a typical real-world problem contains a number of different sub-problems, and that a truly optimal solution (the best combination of cost, performance and feature) is obtained by applying the right tool to the right sub-problem. Here I share some of my perspectives on what constitutes the 'right job' for fuzzy control and describe recent advances in neuro-fuzzy modeling to illustrate and to motivate the synergistic use of different tools.

Chiu, Stephen↗
NASA NTRS1998

The SGI/Cray T3E: Experiences and Insights

The NASA Goddard Space Flight Center is home to the fifth most powerful supercomputer in the world, a 1024 processor SGI/Cray T3E-600. The original 512 processor system was placed at Goddard in March, 1997 as part of a cooperative agreement between the High Performance Computing and Communications Program's Earth and Space Sciences Project (ESS) and SGI/Cray Research. The goal of this system is to facilitate achievement of the Project milestones of 10, 50 and 100 GFLOPS sustained performance on selected Earth and space science application codes. The additional 512 processors were purchased in March, 1998 by the NASA Earth Science Enterprise for the NASA Seasonal to Interannual Prediction Project (NSIPP). These two "halves" still operate as a single system, and must satisfy the unique requirements of both aforementioned groups, as well as guest researchers from the Earth, space, microgravity, manned space flight and aeronautics communities. Few large scalable parallel systems are configured for capability computing, so models are hard to find. This unique environment has created a challenging system administration task, and has yielded some insights into the supercomputing needs of the various NASA Enterprises, as well as insights into the strengths and weaknesses of the T3E architecture and software. The T3E is a distributed memory system in which the processing elements (PE's) are connected by a low latency, high bandwidth bidirectional 3-D torus. Due to the focus on high speed communication between PE's, the T3E requires PE's to be allocated contiguously per job. Further, jobs will only execute on the user specified number of PE's and PE timesharing is possible but impractical. With a highly varied job mix in both size and runtime of jobs, the resulting scenario is PE fragmentation and an inability to achieve near 100% utilization. SGI/Cray has provided several scheduling and configuration tools to minimize the impact of fragmentation. These tools include PScheD (the political scheduler), GRM (the global resource manager) and NQE (the Network Queuing Environment). Features and impact of these tools will be discussed, as will resulting performance and utilization data. As a distributed memory system, the T3E is designed to be programmed through explicit message passing. Consequently, certain assumptions related to code design are made by the operating system (UNICOS/mk) and its scheduling tools. With the exception of HPF, which does run on the T3E, however poorly, alternative programming styles have the potential to impact the T3E in unexpected and undesirable ways. Several examples will be presented (preceeded with the disclaimer, "Don't try this at home! Violators will be prosecuted!")

Bernard, Lisa Hamet↗
NASA NTRS2002

Automated CFD Parameter Studies on Distributed Parallel Computers

The objective of the current work is to build a prototype software system which will automated the process of running CFD jobs on Information Power Grid (IPG) resources. This system should remove the need for user monitoring and intervention of every single CFD job. It should enable the use of many different computers to populate a massive run matrix in the shortest time possible. Such a software system has been developed, and is known as the AeroDB script system. The approach taken for the development of AeroDB was to build several discrete modules. These include a database, a job-launcher module, a run-manager module to monitor each individual job, and a web-based user portal for monitoring of the progress of the parameter study. The details of the design of AeroDB are presented in the following section. The following section provides the results of a parameter study which was performed using AeroDB for the analysis of a reusable launch vehicle (RLV). The paper concludes with a section on the lessons learned in this effort, and ideas for future work in this area.

Rogers, Stuart E.↗
NASA NTRS2003

Learning the ABCs (of Project Management)

To lead a project effectively, one has to establish and maintain the flexibility to take appropriate actions when needed. Overconstrained situations should be avoided. To get on top of matters and stay there, a manager needs to anticipate what it will take to successfully complete the job. Physical and financial resources, personnel, and management structure are all important considerations. Carving out the necessary turf up front can make a world of difference to the project's outcome. After the "what," "where," and "when" of a project are nailed down, the next question is "how" to do the job. When I first interviewed for the job of Science Payload Manager on the Advanced Composition (ACE) Explorer mission, Dr. Edward Stone (ACE Principal Investigator) asked, "Al, give me an idea of your management style." It was a question I had not considered before. I thought about it for a few seconds and then answered, "Well, the first descriptive term that comes to mind is the word "tranquility". That seemed to startle him. So I added, "I guess what I mean is, that if the situation is tranquil and the project is running smoothly, then I've anticipated all the problems and taken necessary actions to head them off." He then asked: "Have you ever reached this state?" "No," I admitted, "but I strive for it." That seemed to satisfy him because I got the job.

Frandsen, Allan↗
NASA NTRS2007

In-Space Crew-Collaborative Task Scheduling

For all past and current human space missions, the final scheduling of tasks to be done in space has been devoid of crew control, flexibility, and insight. Ground controllers, with minimal input from the crew, schedule the tasks and uplink the timeline to the crew or uplink the command sequences to the hardware. Prior to the International Space Station (ISS), the crew could make requests about tomorrow s timeline, they could omit a task, or they could request that something in the timeline be delayed. This lack of control over one's own schedule has had negative consequences. There is anecdotal consensus among astronauts that control over their own schedules will mitigate the stresses of long duration missions. On ISS, a modicum of crew control is provided by the job jar. Ground controllers prepare a task list (a.k.a. "job jar") of non-conflicting tasks from which jobs can be chosen by the in space crew. Because there is little free time and few interesting non-conflicting activities, the task-list approach provides little relief from the tedium of being micro-managed by the timeline. Scheduling for space missions is a complex and laborious undertaking which usually requires a large cadre of trained specialists and suites of complex software tools. It is a giant leap from today s ground prepared timeline (with a job jar) to full crew control of the timeline. However, technological advances, currently in-work or proposed, make it reasonable to consider scheduling a collaborative effort by the ground-based teams and the in-space crew. Collaboration would allow the crew to make minor adjustments, add tasks according to their preferences, understand the reasons for the placement of tasks on the timeline, and provide them a sense of control. In foreseeable but extraordinary situations, such as a quick response to anomalies and extended or unexpected loss of signal, the crew should have the autonomous ability to make appropriate modifications to the timeline, extend the timeline, or even start over with a new timeline. The Vision for Space Exploration (VSE), currently being pursued by the National Aeronautics and Space Administration (NASA), will send humans to Mars in a few decades. Stresses on the human mind will be exacerbated by the longer durations and greater distances, and it will be imperative to implement stress-reducing innovations such as giving the crew control of their daily activities.

Jaap, John↗
NASA NTRS2010

X-Ray Backscatter Machine Support Frame

This summer at Kennedy Space Center, I spent 10 weeks as an intern working at the Prototype Development Lab. During this time I learned about the design and machining done here at NASA. I became familiar with the process from where a design begins in Pro/Engineer and finishes at the hands of the machinists. As an intern I was given various small jobs to do and then one project of my own. My personal project was a job for the Applied Physics Lab; in their work they use an X-Ray Backscatter machine. Previously it was resting atop a temporary frame that limited the use of the machine. My job was to design a frame for the machine to rest upon that would allow a full range of sample sizes. The frame was required to support the machine and provide a strain relief for the cords attached to the machine as it moved in the x and y directions. Calculations also had to be done to be sure the design would be able to withstand any loads or outside sources of stress. After the calculations proved the design to be ready to withstand the requirements, the parts were ordered or fabricated, as required. This helped me understand the full process of jobs sent to the Prototype Development Lab.

Cannon, Brooke↗
NASA NTRS2012

LEGION: Lightweight Expandable Group of Independently Operating Nodes

LEGION is a lightweight C-language software library that enables distributed asynchronous data processing with a loosely coupled set of compute nodes. Loosely coupled means that a node can offer itself in service to a larger task at any time and can withdraw itself from service at any time, provided it is not actively engaged in an assignment. The main program, i.e., the one attempting to solve the larger task, does not need to know up front which nodes will be available, how many nodes will be available, or at what times the nodes will be available, which is normally the case in a "volunteer computing" framework. The LEGION software accomplishes its goals by providing message-based, inter-process communication similar to MPI (message passing interface), but without the tight coupling requirements. The software is lightweight and easy to install as it is written in standard C with no exotic library dependencies. LEGION has been demonstrated in a challenging planetary science application in which a machine learning system is used in closed-loop fashion to efficiently explore the input parameter space of a complex numerical simulation. The machine learning system decides which jobs to run through the simulator; then, through LEGION calls, the system farms those jobs out to a collection of compute nodes, retrieves the job results as they become available, and updates a predictive model of how the simulator maps inputs to outputs. The machine learning system decides which new set of jobs would be most informative to run given the results so far; this basic loop is repeated until sufficient insight into the physical system modeled by the simulator is obtained.

Burl, Michael C.↗