Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Task 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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2017

Addressing Global Data Dependencies in Heterogeneous Asynchronous Runtime Systems on GPUs

Large-scale parallel applications with complex global data dependencies beyond those of reductions pose significant scalability challenges in an asynchronous runtime system. Internodal challenges include identifying the all-to-all communication of data dependencies among the nodes. Intranodal challenges include gathering together these data dependencies into usable data objects while avoiding data duplication. This paper addresses these challenges within the context of a large-scale, industrial coal boiler simulation using the Uintah asynchronous many-task runtime system on GPU architectures. We show significant reduction in time spent analyzing data dependencies through refinements in our dependency search algorithm. Multiple task graphs are used to eliminate subsequent analysis when task graphs change in predictable and repeatable ways. Using a combined data store and task scheduler redesign reduces data dependency duplication ensuring that problems fit within host and GPU memory. Furthermore, these modifications did not require any changes to application code or sweeping changes to the Uintah runtime system. We report results running on the DOE Titan system on 119K CPU cores and 7.5K GPUs simultaneously. Our solutions can be generalized to other task dependency problems with global dependencies among thousands of nodes which must be processed efficiently at large scale.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

An automatic performance model-based scheduling tool for coupled climate system models

The prediction of the climate system is highly depended on the efficient integration of observations and simulations of the Earth. This is regarded as a canonical example of the cyber-physical system. The climate system model, the simulation engine in this cyber-physical system, is one of most challenging applications in scientific computing. It utilizes the multi-physics simulation that couples multiple components, conducts decadal to millennium simulations, and has long been an important application on supercomputers. However, current climate system models suffer from the inefficient task scheduling methods resulting in an intolerable simulation time. Take the Community Earth System Model (CESM), the most widely used climate system model, as an example, one major reason that CESM suffers from bad performances is the huge overhead to rationally distribute processes among the coupled heterogeneous components. According to the report of NCAR, every percent improvement in CESM performance frees up to the equivalent of $250,000 in computing resources in their scientific experiments. To address such challenge, our paper firstly constructs a lightweight and accurate performance model for effectively capturing and predicting the heterogeneous time-to-solution performance of end-to-end CESM components with a given simulation configuration. Then, based on the performance model, we further propose an efficient scheduling strategy based on rectangular packing method to determine the best process layout among different components, and the process numbers assigned to each component. Our evaluations show that we can achieve 58% average run time reductions on CESM comparing to the widely used sequential process layout for a scale of 144 to 480 cores on typical CPU clusters. And we can save 4 million CPU hours when we conduct one standard scientific experiment (a 2870-year simulation), which equals to save $40,089 with a charge of $0.01 per CPU hour. Meanwhile, 26% extra performance improvements could also be gained in our methods comparing to the heuristic branch and bound algorithm with the guidance of the known curve-fitting performance model.

58 GEOSCIENCES↗
DOE OSTI2020

Remote Radiation Sensing Using Aerial and Ground Platforms

Remote sensing of ionizing radiation has a significant role in waste management, nuclear material management and nonproliferation, and radiation safety. Robotic platforms can surpass the number of tasks that are achieved by humans. With this technique, the operator's radiation exposure can be decreased. Remote sensing allows for the evaluation and monitoring of radiological contamination. Gamma-ray and neutron sensors were integrated onto the robotic platforms. This approach allows for the radiation sensor data to be dynamically tracked and mapped thus enabling further analysis of the radiation flux in temporal and spatial domains. The goal is to complete scheduled tasks while the robot is being irradiated. To achieve this, electronic components must be shielded and radiation hardened. CZT Detector: Cadmium Zinc Telluride (CZT) detector technology has been a promising solution for gamma-ray and x-ray measurements. Detector data is transferred to the Odroid minicomputer that controls and powers the module via the USB. Robot Operating System (ROS) was utilized for data acquisition and data fusion. The Mariscotti method was employed for the spectrum analysis. A function was programmed in ROS for the automatic identification of photopeaks. CLYC Detector: A Cs{sub 2}LiYCl{sub 6}:Ce{sup 3+} (CLYC) detector was used for simultaneous medium-resolution gamma-ray measurements and neutron counting. A 2.54 cm diameter photomultiplier tube (PMT) was equipped with a high voltage supply and a miniature digitizer. Gamma-ray excitation: fast core-to-valence luminescence (CVL) with 1 ns decay constant, and prompt Ce{sup 3+} emission with 50 ns decay constant. Neutron excitation: slow cerium self-trapped excitation (Ce{sup 3+} STE), 1000 ns decay constant. Radiation Source Localization: Maximum Likelihood Estimation (MLE) and gradient-based methods were used to locate the position of a radiation source based on measured radiation intensities. Multi-Particle Transport Code FLUKA: Estimation of radiation damage of the electronic components is important in order to optimize the robot's operational time while it is irradiated. Displacement per atom (DPA) represents the radiation damage in materials exposed to the ionizing radiation. Various shielding layers of different thickness t were analyzed (< 5% statistical error). The model of the controller of the UAS was designed in FLUKA. Conclusion: CZT and CLYC detectors were integrated onto the robotic platforms. Radiation source localization and contour mapping using robotic platforms were studied. Functions for data analysis and fusion were developed in ROS. FLUKA code was utilized to analyze DPA values. Layers of low-density and high-density materials were used to shield the UAS electronics.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2019

Topology-aware task-assIGnment mappER (TIGER) v0.1

The general topology-aware task assignment problem is an NP-complete graph isomorphism problem. Existing task scheduling approaches are either heuristic or based on physical optimization algorithms, providing different speed and solution quality trade-offs. Physical Ising machines such as quantum and digital annealers have recently become available offering an alternative hardware solution to solve certain types of optimization problems. We propose an algorithm that allows expressing the problem in the compatible format. We present a domain specific partition strategy that enables to execute larger scale problems. TIGER - topology-aware task assignment mapper tool - implements the proposed algorithm and automatically integrates task- communication graph and an architecture graph into the quantum software environment.

Donofrio, David↗
DOE OSTI2022

Implementation of a practical Markov chain Monte Carlo sampling algorithm in PyBioNetFit

Abstract Summary Bayesian inference in biological modeling commonly relies on Markov chain Monte Carlo (MCMC) sampling of a multidimensional and non-Gaussian posterior distribution that is not analytically tractable. Here, we present the implementation of a practical MCMC method in the open-source software package PyBioNetFit (PyBNF), which is designed to support parameterization of mathematical models for biological systems. The new MCMC method, am, incorporates an adaptive move proposal distribution. For warm starts, sampling can be initiated at a specified location in parameter space and with a multivariate Gaussian proposal distribution defined initially by a specified covariance matrix. Multiple chains can be generated in parallel using a computer cluster. We demonstrate that am can be used to successfully solve real-world Bayesian inference problems, including forecasting of new Coronavirus Disease 2019 case detection with Bayesian quantification of forecast uncertainty. Availability and implementation PyBNF version 1.1.9, the first stable release with am, is available at PyPI and can be installed using the pip package-management system on platforms that have a working installation of Python 3. PyBNF relies on libRoadRunner and BioNetGen for simulations (e.g. numerical integration of ordinary differential equations defined in SBML or BNGL files) and Dask.Distributed for task scheduling on Linux computer clusters. The Python source code can be freely downloaded/cloned from GitHub and used and modified under terms of the BSD-3 license (https://github.com/lanl/pybnf). Online documentation covering installation/usage is available (https://pybnf.readthedocs.io/en/latest/). A tutorial video is available on YouTube (https://www.youtube.com/watch?v=2aRqpqFOiS4&t=63s). Supplementary information Supplementary data are available at Bioinformatics online.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Kensor: Coordinated Intelligence from Co-Located Sensors

Internet of Things (IoT) is becoming more pervasive in many installations, including homes, manufacturing plants, and industrial facilities of all kinds. The data that IoT produces is a reflection of usual behavior such as daily routines and scheduled tasks, but also from unexpected behavior due to unintentional or undesirable abnormalities. Here, we focus on achieving coordinated intelligence about normal and abnormal phenomena from multiple sensors that are geographically colocated in close proximity, monitoring and controlling a set of co-located devices. Given a set of co-located sensors, we seek an intelligent approach that would automatically determine the “normal” patterns of behaviors among the correlated sensors. After normal behavior is extracted, later monitoring should detect any deviant variations over time. An example application is an entry monitoring and alert system for facilities such as nuclear reactors, where badge readers, door locks, lights, weight trackers and other co-located sensors at the entry point are collectively tracked. To address this problem, we identify the possible solution approach that can be used to solve its different variants. The implemented model is developed as a combination of rules and Markov Chain methods.

Kotevska, Olivera↗
DOE OSTI2022

Exploring the Use of Novel Spatial Accelerators in Scientific Applications

Driven by the need to find alternative accelerators which can viably replace GPUs in next-generation Supercomputing systems, this paper proposes a methodology to enable agile application/hardware co-design. The application-first methodology provides the ability to come up with design of accelerators while working with real-world workloads, available accelerators, and system software. The iterative design process targets a set of kernels in a workload for performance estimates that can prune the design space for later phases of detailed architectural evaluations. To this effect, in this paper, a novel data-parallel device model is introduced that simulates the latency of performance-sensitive operations in an accelerator including data transfers and kernel computation using multi-core CPUs. The use of off-the-shelf simulators, such as pre-RTL simulator Aladdin or multiple tools available for exploring the design of deep neural network accelerators (e.g., Timeloop) is demonstrated for evaluation of various accelerator designs using applications with realistic inputs. Examples of multiple device configurations that are instantiable in a system are explored to evaluate the performance benefit of deploying novel accelerators. The proposed device is integrated with a programming model and system software to potentially explore the impacts of high-level programming languages/compilers and low-level effects such as task scheduling on multiple accelerators. We analyze our methodology for a set of applications that represent high-performance computing (HPC) and graph analytics. The applications include a computational chemistry kernel realized using tensor contractions, triangle counting, GraphSAGE and Breadth-first Search. These applications include kernels such as dense matrix-dense matrix multiplication, sparse matrix-spare matrix multiplication, and sparse matrix-dense vector multiplication. Our results indicate potential performance benefits and insights for system design by including accelerators that realize these kernels along-side general purpose accelerators.

AI, codesign, Accelerated Computing, Modeling and ↗
DOE OSTI2021

`SkyPy`: A package for modelling the Universe

SkyPy is an open-source Python package for simulating the astrophysical sky. It comprises a library of physical and empirical models across a range of observables and a command-line script to run end-to-end simulations. The library provides functions that sample realisations of sources and their associated properties from probability distributions. Simulation pipelines are constructed from these models using a YAML-based configuration syntax, while task scheduling and data dependencies are handled internally and the modular design allows users to interface with external software. SkyPy is developed and maintained by a diverse community of domain experts with a focus on software sustainability and interoperability. By fostering development, it provides a framework for correlated simulations of a range of cosmological probes including galaxy populations, large scale structure, the cosmic microwave background, supernovae and gravitational waves. Version 0.4 implements functions that model various properties of galaxies including luminosity functions, redshift distributions and optical photometry from spectral energy distribution templates. Future releases will provide additional modules, for example, to simulate populations of dark matter halos and model the galaxy-halo connection, making use of existing software packages from the astrophysics community where appropriate.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2021

Evaluation of the Los Alamos Nuclear Material Packaging Risk Ranking Method

Repackaging nuclear material into robust containers to protect workers and the public has been ongoing at LANL and around the DOE complex for nearly two decades. The number of containers at LANL is around 5,000; limited resources for repackaging material has led to extended repackaging campaigns and the need to prioritize repackaging. Various methodologies have been used to prioritize the repackaging efforts and to demonstrate progress in risk reduction over time (e.g., Boerigter, 1997). The 2000-1 DNFSB recommendation recognized the limited DOE resources for repackaging, and acknowledged the need to “prioritize and schedule tasks to be undertaken with available funds according to consideration of risks.” Later, in DNFSB recommendation 2005-1, in addition to recommending that DOE develop a packaging standard, the Board recommended that “Characterization information should also be used to develop a surveillance program prioritized according to expected material and container risk (including, for example, material type, material form, and the age and type of container).” In response to requests and recommendations from the DOE and DNSFB to prioritize according to worker risk, a risk ranking method based on the potential consequence of dropping a container from 3 meters was developed in 2007 (Smith, 2007) and updated in 2014 (Hoffman, 2014). Various LANL implementation plans for repackaging were developed over the years using this methodology (Stone, 2014). Currently, this method is utilized in conjunction with an algorithm to mitigate programmatic risk to prioritize container repackaging and material processing (Prochnow, 2015). The purpose of this study is to document how the current risk ranking method works, how it is used and potential limitations.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2022

The Early Career Research Program Annual Progress Report: Using ARM Data and Multiscale Models to Advance the Understanding of Liquid-Phase Cloud Response to Aerosol Perturbation over Ocean and Land

This report is for the first fiscal year of this ECRP project. This ECRP project aims to study the response of liquid-phase clouds to aerosol perturbation and reduce the related uncertainty in the DOE Earth system model. During this fiscal year, this project mainly made progress on the tasks about observational analyses for Objective 1; and successfully set up two Large-Eddy Simulation (LES) models for Objective 2. A postdoctoral researcher and a graduate student summer intern are on board and fully dedicated to this project. In addition to internal collaboration within LLNL, this project has routinely maintained external collaboration with scientists from universities, NASA Langley Research Center, and other DOE national laboratories. For the next reporting period, we plan to complete the tasks scheduled for the second year. Furthermore, we will promote the findings to communities of interest and proactively seek new collaborations to increase the impact of this project.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Systems and methods for tensor scheduling

A technique for efficient scheduling of operations in a program for parallelized execution thereof using a multi-processor runtime environment having two or more processors includes constraining the type or number of loop optimization transforms that may be explored such that memory and processing capacity available for the scheduling task are not exceeded, while facilitating a tradeoff between memory locality, parallelization, and/or data communication between memory modules of the multi-processor runtime environment.

Meister, Benoit J.↗
DOE OSTI2018

Schedule Management Optimization (SMO) Domain Model (V.1.2)

Schedule Management Optimization (SMO) is a tool for automatically generating a schedule of project tasks. Project scheduling is traditionally achieved with the use of commercial project management software or case-specific optimization formulations. Commercial software packages are useful tools for managing and visualizing copious amounts of project task data. However, their ability to automatically generate optimized schedules is limited. Furthermore, there are many real-world constraints and decision variables that commercial packages ignore. Case-specific optimization formulations effectively identify schedules that optimize one or more objectives for a specific problem, but they are unable to handle a diverse selection of scheduling problems. SMO enables practitioners to generate optimal project schedules automatically while considering a broad range of real-world problem characteristics. SMO has been designed to handle some of the most difficult scheduling problems – those with resource constraints, multiple objectives, multiple inventories, and diverse ways of performing tasks. This report contains descriptions of the SMO modeling concepts and explains how they map to real-world scheduling considerations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

An Approximation Algorithm for a Task Allocation, Sequencing and Scheduling Problem Involving a Human-Robot Team

Here we present an approximation algorithm for a Task Allocation, Sequencing and Scheduling Problem (TASSP) involving a team of human operators and robots. The robots have to travel to a given set of targets and collaboratively work on the tasks at the targets with the human operators. The problem aims to find a sequence of targets for each robot to visit and schedule the tasks at the targets with the human operators such that each target is visited exactly once by some robot, the scheduling constraints are satisfied and the maximum mission time of any robot is minimum. This problem is a generalization of the single Traveling Salesman Problem and is NP-Hard. Given k robots and m human operators, an algorithm is developed for solving the TASSP with an approximation ratio equal to 5/2- 1/k when m ≥ k and equal to 7/2 -1/k otherwise. Computational results are also presented to corroborate the performance of the proposed algorithm.

42 ENGINEERING↗
DOE OSTI2017

Aluminum Clad Spent Nuclear Fuel Long Term Dry Storage Technical Issues Action Plan - Technical and Engineering Activities

The DOE Spent Nuclear Fuel Working Group (SNFWG)’s report on technical considerations and challenges for extended dry storage of aluminum-clad spent nuclear fuel (ASNF) entitled, Aluminum-Clad Spent Nuclear Fuel: Technical Consideration and Challenges for Extended (>50 Years) Dry Storage, DOE/ID RPT-1575, June 2017 identified five knowledge gaps and technical data needs. This DOE SNFWG report also made several recommendations including one to develop an action plan to identify needed technical and engineering activities and analyses to address the identified knowledge gaps and technical data needs. A team composed of twenty SNF management experts and appropriate subject matter experts from the Idaho National Laboratory and Savannah River National Laboratory was assembled to develop this action plan. The action plan identifies and recommends six technical and engineering tasks and estimated cost and schedule for task completion. It is estimated that $5M and 18 months will be required to execute the recommended tasks.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2018

Performance Tradeoffs of Spectrum Sensing and Target State Estimation and Fusion

We consider a scenario where a cognitive radio sensor performs spectrum sensing and target tracking alternately. Both tasks may impose certain performance guarantee requirements that necessitate considerations of how to schedule these tasks over time. In addition, multiple such cognitive radio sensors can perform collaborative spectrum sensing and target tracking, where a fusion center serves to collect the sensor estimates and their covariances to generate fused target state estimates. We study the performance tradeoffs of these two tasks from variable scheduling intervals. The effect of sensing errors on the target tracking performance due to induced loss from perceived channel unavailability is investigated and conclusions are drawn based on the observed performance tradeoffs.

Liu, Qiang↗
DOE OSTI2024

IRIS Reimagined: Advancements in Intelligent Runtime System for Task-Based Programming

Task-based programming models are gaining traction in scientific computing. IRIS is a portable runtime system that exploits multiple heterogeneous programming systems and can discover available resources and manage multiple diverse programming systems (e.g., CUDA, Hexagon, HIP, Level Zero, OpenCL, and OpenMP) simultaneously. It accounts for the constraints of task dependencies and provides customizable scheduling policies to map those tasks to heterogeneous devices. In this paper, we present new capabilities added to IRIS to improve its portability for heterogeneous programming, build-friendliness, and performance efficiency. The new additions include vendor-specific kernel support, a runtime system with a foreign function interface to eliminate writing wrapper or boilerplate code for heterogeneous kernels, an easy-to-use and configurable CMake-based build environment, automatic and efficient data transfers and orchestration, and the Hunter and DAGGER toolchains to evaluate IRIS’s task scheduling algorithms.

Miniskar, Narasinga Rao↗
DOE OSTI2020

Characterizing the Performance of Executing Many-tasks on Summit

Many scientific workloads are comprised of many tasks, where each task is an independent simulation or analysis of data. The execution of millions of tasks on heterogeneous HPC platforms requires scalable dynamic resource management and multi-level scheduling. RADICAL-Pilot (RP) -- an implementation of the Pilot abstraction, addresses these challenges and serves as an effective runtime system to execute workloads comprised of many tasks. In this paper, we characterize the performance of executing many tasks using RP when interfaced with JSM and PRRTE on Summit: RP is responsible for resource management and task scheduling on acquired resource; JSM or PRRTE enact the placement of launching of scheduled tasks. Our experiments provide lower bounds on the performance of RP when integrated with JSM and PRRTE. Specifically, for workloads comprised of homogeneous single-core, 15 minutes-long tasks we find that: PRRTE scales better than JSM for > O(1000) tasks; PRRTE overheads are negligible; and PRRTE supports optimizations that lower the impact of overheads and enable resource utilization of 63% when executing O(16K), 1-core tasks over 404 compute nodes.

Matteo, Turilli↗