Search NASA⌕ Search

SEARCH · Search NASA

Results for “Communication aggregation”

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 73 records · Page 4

Designing Scalable PGAS Communication Subsystems on Cray Gemini Interconnect

The Cray Gemini Interconnect has been recently introduced as a next generation network architecture for building multi-petaflop supercomputers. Cray XE6 systems including LANL Cielo, NERSC Hopper, ORNL Titan and proposed NCSA BlueWaters leverage the Gemini Interconnect as their primary Interconnection network. At the same time, programming models such as the Message Passing Interface (MPI) and Partitioned Global Address Space (PGAS) models such as Unified Parallel C (UPC) and Co-Array Fortran (CAF) have become available on these systems. Global Arrays is a popular PGAS model used in a variety of application domains including hydrodynamics, chemistry and visualization. Global Arrays uses Aggregate Re- mote Memory Copy Interface (ARMCI) as the communication runtime system for Remote Memory Access communication. This paper presents a design, implementation and performance evaluation of scalable and high performance communication subsystems on Cray Gemini Interconnect using ARMCI. The design space is explored and time-space complexities of commu- nication protocols for one-sided communication primitives such as contiguous and uniformly non-contiguous datatypes, atomic memory operations (AMOs) and memory synchronization is presented. An implementation of the proposed design (referred as ARMCI-Gemini) demonstrates the efficacy on communication primitives, application kernels such as LU decomposition and full applications such as Smooth Particle Hydrodynamics (SPH) application.

ARMCI↗

Global‐view coefficients: a data management solution for parallel quantum Monte Carlo applications

Summary Quantum Monte Carlo (QMC) applications perform simulation with respect to an initial state of the quantum mechanical system, which is often captured by using a cubic B‐spline basis. This representation is stored as a read‐only table of coefficients and accesses to the table are generated at random as part of the Monte Carlo simulation. Current QMC applications, such as QWalk and QMCPACK, replicate this table at every process or node, which limits scalability because increasing the number of processors does not enable larger systems to be run. We present a partitioned global address space approach to transparently managing this data using Global Arrays in a manner that allows the memory of multiple nodes to be aggregated. We develop an automated data management system that significantly reduces communication overheads, enabling new capabilities for QMC codes. Experimental results with QWalk and QMCPACK demonstrate the effectiveness of the data management system. Copyright © 2016 John Wiley & Sons, Ltd.

Niu, Qingpeng↗

Large-Signal Stability of Phase-Balanced Equilibria in Single-Phase Grid-Forming Inverter Systems

This article explores the setup where large numbers of single-phase grid-forming inverters with droop control across distribution networks self-organize into a stable and balanced system with 120° phase offsets across aggregates in the absence of balanced three-phase generating resources or external communication. A suite of circuit- and system-theoretic notions are leveraged to derive a dynamical model for phase-angle differences across aggregates of inverters connected in the three phases. Focusing on this model, large-signal stability is established and the region of attraction of the phase-balanced equilibria is determined with the aid of a Lyapunov function. Experimental validation for a bench-top prototype network is included to support the analytical developments. Altogether, the effort supports the vision of facilitating balanced operation of distribution networks with grid-forming inverters during service disruptions at the bulk transmission network.

Droop control↗

Bridging Control and Deployment: A Cross-Layer Analysis of Scalable Building Cluster Control

Building cluster control has emerged as a promising approach for enabling flexible and coordinated operation of distributed building systems, yet its transition from pilot demonstrations to routine grid-interactive operation remains limited. This paper argues that this gap cannot be explained by control algorithms alone. Instead, it arises from interacting barriers in communication infrastructure, data and semantic interoperability, uncertainty management, stakeholder participation, market design, and policy support. Accordingly, the paper reviews both technical and non-technical barriers to building cluster control. Technical challenges include heterogeneous devices and protocols, communication latency and reliability, distributed decision-making, and uncertainty propagation across aggregated loads. Non-technical barriers include user participation, stakeholder coordination, incentive allocation, and data governance. Existing solution approaches are synthesized, including semantic interoperability frameworks, edge and hierarchical communication architectures, distributed and transactive control strategies, uncertainty-aware optimization, policy mechanisms, and market reforms. Based on this analysis, two research directions are identified: testing infrastructures that can evaluate control performance under realistic multi-building conditions, and abstraction methods that allow building clusters to interact with other energy sectors through standardized flexibility representations. Overall, the paper provides a structured review of how building cluster control can move from isolated demonstrations toward reproducible, market-compatible, and grid-relevant implementation.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Havery Mudd 2014-2015 Computer Science Conduit Clinic. Final Report

Conduit, a new open-source library developed at Lawrence Livermore National Laboratories, provides a C++ application programming interface (API) to describe and access scientific data. Conduit’s primary use is for inmemory data exchange in high performance computing (HPC) applications. Our team tested and improved Conduit to make it more appealing to potential adopters in the HPC community. We extended Conduit’s capabilities by prototyping four libraries: one for parallel communication using MPI, one for I/O functionality, one for aggregating performance data, and one for data visualization.

97 MATHEMATICS AND COMPUTING↗

Design and Evaluation of a Secure Virtual Power Plant.

For three years, Sandia National Laboratories, Georgia Institute of Technology, and University of Illinois at Urbana-Champaign investigated a smart grid vision in which renewable-centric Virtual Power Plants (VPPs) provided ancillary services with interoperable distributed energy resources (DER). This team researched, designed, built, and evaluated real-time VPP designs incorporating DER forecasting, stochastic optimization, controls, and cyber security to construct a system capable of delivering reliable ancillary services, which have been traditionally provided by large power plants or other dedicated equipment. VPPs have become possible through an evolving landscape of state and national interconnection standards, which now require DER to include grid-support functionality and communications capabilities. This makes it possible for third party aggregators to provide a range of critical grid services such as voltage regulation, frequency regulation, and contingency reserves to grid operators. This paradigm (a) enables renewable energy, demand response, and energy storage to participate in grid operations and provide grid services, (b) improves grid reliability by providing additional operating reserves for utilities, independent system operators (ISOs), and regional transmission organization (RTOs), and (c) removes renewable energy high-penetration barriers by providing services with photovoltaics and wind resources that traditionally were the jobs of thermal generators. Therefore, it is believed VPP deployment will have far-reaching positive consequences for grid operations and may provide a robust pathway to high penetrations of renewables on US power systems. In this report, we design VPPs to provide a range of grid-support services and demonstrate one VPP which simultaneously provides bulk-system energy and ancillary reserves.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Interactive Buildings: A Review

Buildings are widely regarded as potential sources for demand flexibility. The flexibility of thermal and electric load in buildings is a result of their interactive nature and its impact on the building’s performance. In this paper, the interaction of a building with the three interaction counterparts of the physical environment, civil infrastructure networks and other buildings is investigated. The literature review presents a wide variety of pathways of interaction and their associated potential impacts on building performance metrics such as net energy use, emissions, occupant comfort and operational cost. It is demonstrated that all of these counterparts of interaction should be considered to harness the flexibility potential of the buildings while maintaining other buildings performance metrics at a desired level. Juxtaposed with the upside potential for providing demand flexibility, numerous implementation challenges are identified that are associated with the evaluation and financial valuation of the capacity for demand flexibility, the aggregated flexibility potential, as well as the control and communication to facilitate the interactions.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Anomaly Detection in Dynamic Networks

Anomaly detection in dynamic communication networks has many important security applications. These networks can be extremely large and so detecting any changes in their structure can be computationally challenging; hence, computationally fast, parallelisable methods for monitoring the network are paramount. For this reason the methods presented here use independent node and edge based models to detect locally anomalous substructures within communication networks. As a first stage, the aim is to detect changes in the data streams arising from node or edge communications. Throughout the thesis simple, conjugate Bayesian models for counting processes are used to model these data streams. A second stage of analysis can then be performed on a much reduced subset of the network comprising nodes and edges which have been identified as potentially anomalous in the first stage. The first method assumes communications in a network arise from an inhomogeneous Poisson process with piecewise constant intensity. Anomaly detection is then treated as a changepoint problem on the intensities. The changepoint model is extended to incorporate seasonal behavior inherent in communication networks. This seasonal behavior is also viewed as a changepoint problem acting on a piecewise constant Poisson process. In a static time frame, inference is made on this extended model via a Gibbs sampling strategy. In a sequential time frame, where the data arrive as a stream, a novel, fast Sequential Monte Carlo (SMC) algorithm is introduced to sample from the sequence of posterior distributions of the change points over time. A second method is considered for monitoring communications in a large scale computer network. The usage patterns in these types of networks are very bursty in nature and don’t fit a Poisson process model. For tractable inference, discrete time models are considered, where the data are aggregated into discrete time periods and probability models are fitted to the communication counts. In a sequential analysis, anomalous behavior is then identified from outlying behavior with respect to the fitted predictive probability models. Seasonality is again incorporated into the model and is treated as a changepoint model on the transition probabilities of a discrete time Markov process. Second stage analytics are then developed which combine anomalous edges to identify anomalous substructures in the network.

97 MATHEMATICS AND COMPUTING↗

Data Intensive Architecture for Scalable Cyber Analytics

Cyber analysts are tasked with the identification and mitigation of network exploits and threats. These compromises are difficult to identify due to the characteristics of cyber communication, the volume of traffic, and the duration of possible attack. In this paper, we describe a prototype implementation designed to provide cyber analysts an environment where they can interactively explore a month’s worth of cyber security data. This prototype utilized On-Line Analytical Processing (OLAP) techniques to present a data cube to the analysts. The cube provides a summary of the data, allowing trends to be easily identified as well as the ability to easily pull up the original records comprising an event of interest. The cube was built using SQL Server Analysis Services (SSAS), with the interface to the cube provided by Tableau. This software infrastructure was supported by a novel hardware architecture comprising a Netezza TwinFin® for the underlying data warehouse and a cube server with a FusionIO drive hosting the data cube. We evaluated this environment on a month’s worth of artificial, but realistic, data using multiple queries provided by our cyber analysts. As our results indicate, OLAP technology has progressed to the point where it is in a unique position to provide novel insights to cyber analysts, as long as it is supported by an appropriate data intensive architecture.

Olsen, Bryan K.↗

Field Validation of a Grid-Interactive Efficient Building Software Solution

The U.S. General Services Administration's (GSA's) Green Proving Ground (GPG) program, in partnership with the National Laboratory of the Rockies (NLR), completed a field study of a Grid-Interactive Efficient Buildings (GEB) software solution. The study focused on a single testbed facility to test the GEB functionality of the software solution, along with other features. The testbed facility - a courthouse - is a common building type in GSA's vast building portfolio, offering potentially impactful findings on a scalable level. The study evaluated Prescriptive Data's technology, Nantum OS, a connected building operating system ("GEB Solution") which aggregates multiple sources of previously siloed building data and combines that data with external sources, such as weather information or utility signals, into a single integrated platform. A GEB Solution is a type of Energy Management Information System (EMIS). EMIS is defined as a system of devices, data services, and software applications that communicates with any building system or third-party data source to aggregate and transform data into new capabilities to aid in the optimization of energy use at the building, campus, or agency level. This specific GEB Solution is an EMIS with ASO, automated system optimization, offering supervisory control of certain aspects of the Building Automation System (BAS). Multiple features were evaluated including, but not limited to, Continuous Demand Management to avoid setting new monthly kilowatt (kW) peaks, energy efficiency for reduction of kilowatt hours (kWh) and natural gas consumption, and automated demand response (ADR) for purposes of lowering demand during a utility called Demand Response (DR) event. The testbed facility was the Foley Federal Building and US Courthouse ("Foley Federal Building") located in Las Vegas, NV. This is a 209,496 sq. ft. building constructed in the 1960s with major renovations in 2004. The facility was a good candidate due to the large prevalence of office and courthouse spaces in the GSA portfolio of buildings. It also has many features which allow integration into and control of the building and a strong facilities team to assist with the study. Quantitative and qualitative performance objectives were developed using GSA's GPG GEB project template along with input from the vendor and building facility staff; these are outlined in Table 1. The quantitative performance objectives focused on continuous demand management, energy efficiency, and automated demand response. The qualitative performance objectives focused on the ease of installation and commissioning as well as the operability of the GEB solution. Other performance metrics that are reported on include carbon reduction, cost effectiveness, and occupant acceptance.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

An evaluation of the CORAL interconnects

The US Department of Energy deployed the Summit and Sierra supercomputers with the latest state-of-the-art network interconnect technology in 2018 and both systems entered production in 2019. In this paper, we provide an in-depth assessment of the systems' network interconnects that are based on Enhanced Data Rate (EDR) 100 Gb/s Mellanox InfiniBand. Both systems use second-generation EDR Host Channel Adapters (HCAs) and switches with several new features such as Adaptive Routing (AR), switch-based collectives, and HCA-based tag matching. Although based on the same components, Summit's network is "non-blocking" (i.e., a fully provisioned Clos network) and Sierra's network has a 2:1 taper between the racks and aggregation switches. We evaluate the two systems' interconnects using traditional communication benchmarks as well as production applications. We find that the new Adaptive Routing dramatically improves performance but the other new features still need improvement.

Zimmer, Christopher↗

Development of a Configurable DERMS Test System in GridAPPS-D

Distributed Energy Resources and their increased penetration within the power grid are a suitable method to provide essential reliability services. In order to do so, they must operate in aggregate. Aggregation of a wide variety of systems with varying standards and protocols is a complex task, which could benefit from rules and standards to support interoperability. To this end, the Portland State University Power Lab has developed the Energy Grid of Things project, which includes a Distributed Energy Resource Management System to provide aggregation and demonstrate the usage of an Energy Services Interface: a set of rules governing the interface between resource owners and aggregator businesses. This thesis presents a Modeling Environment that provides a simulation and test platform reflecting the complex needs of an aggregator. A grid simulator is required that is able to reflect the effects of thousands of Distributed Energy Resources operating in tandem; furthermore, these effects must be communicated to external actors such as the Grid Operator, which provides grid service requests to the aggregator and requires feedback. The Modeling Environment provides all of this functionality by leveraging the GridAPPS-D platform to develop an interoperable class-based system that provides configurable simulations with modular input and output APIs. The Modeling Environment uses abstractions to allow any feasible type of input to be reflected within the model as a Distributed Energy Resource, and serves as a proof-of-concept for the development of aggregation simulation and testing within GridAPPS-D.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Frequency Regulation Services from Connected Residential Devices: Short Paper

In this paper, we demonstrate the potential benefits that residential buildings can provide for frequency regulation services in the electric power grid. In a hardware-in-the- loop (HIL) implementation, simulated homes along with a physical laboratory home are coordinated via a grid aggregator, and it is shown that their aggregate response has the potential to follow the regulation signal on a timescale of seconds. Connected (communication-enabled), devices in the National Renewable Energy Laboratory's (NREL's) Energy Systems Integration Facility (ESIF) received demand response (DR) requests from a grid aggregator, and the devices responded accordingly to meet the signal while satisfying user comfort bounds and physical hardware limitations. Future research will address the issues of cybersecurity threats, participation rates, and reducing equipment wear-and-tear while providing grid services.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Grid Connected Functionality

Dataset demonstrating the potential benefits that residential buildings can provide for frequency regulation services in the electric power grid. In a hardware-in-the-loop (HIL) implementation, simulated homes along with a physical laboratory home are coordinated via a grid aggregator, and it is shown that their aggregate response has the potential to follow the regulation signal on a timescale of seconds. Connected (communication-enabled), devices in the National Renewable Energy Laboratory's (NREL's) Energy Systems Integration Facility (ESIF) received demand response (DR) requests from a grid aggregator, and the devices responded accordingly to meet the signal while satisfying user comfort bounds and physical hardware limitations.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Short Paper: Frequency Regulation Services from Connected Residential Devices: Preprint

In this paper, we demonstrate the potential benefits that residential buildings can provide for frequency regulation services in the electric power grid. In a hardware-in-the- loop (HIL) implementation, simulated homes along with a physical laboratory home are coordinated via a grid aggregator, and it is shown that their aggregate response has the potential to follow the regulation signal on a timescale of seconds. Connected (communication-enabled), devices in the National Renewable Energy Laboratory's (NREL's) Energy Systems Integration Facility (ESIF) received demand response (DR) requests from a grid aggregator, and the devices responded accordingly to meet the signal while satisfying user comfort bounds and physical hardware limitations. Future research will address the issues of cybersecurity threats, participation rates, and reducing equipment wear-and-tear while providing grid services.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Game-Theoretic Approach for Electricity Pricing Between Distribution System Operator and Load Aggregators

Transactive energy has emerged and gained more and more attention in recent years along with smart grid two-way communication technologies. Electricity pricing strategies play a critical role in influencing and shaping the customers’ energy load profile. In this paper, we investigate the electricity pricing strategy between a distribution system operator (DSO) and load aggregators (LAs) by adopting a bi-level Stackelberg game approach. With the purpose of maximizing its own operating revenue as well as better serving its customers, the upper level DSO will determine the price considering how lower level LAs will respond to it. Peak-to-average ratio of the total demand is modeled as a constraint and peak load is penalized at the upper optimization level. Classical backward induction is used to obtain a Nash equilibrium for the formulated bi-level Stackelberg game. Numerical results demonstrate the effectiveness of the proposed game approach in leveraging flexible demand potential to benefit both DSO and LAs.

Chen, Yang↗

RuralAI in Tomato Farming: Integrated Sensor System, Distributed Computing, and Hierarchical Federated Learning for Crop Health Monitoring

Precision horticulture is evolving due to scalable sensor deployment and machine learning (ML) integration. These advancements boost the operational efficiency of individual farms, balancing the benefits of analytics with autonomy requirements. However, given concerns that affect wide geographic regions (e.g., climate change), there is a need to apply models that span farms. Federated learning (FL) has emerged as a potential solution. FL enables decentralized ML across different farms without sharing private data. Traditional FL assumes simple two-tier network topologies and, thus, falls short of operating on more complex networks found in real-world agricultural scenarios. Networks vary across crops and farms and encompass various sensor data modes, extending across jurisdictions. New hierarchical FL (HFL) approaches are needed for more efficient and context-sensitive model sharing, accommodating regulations across multiple jurisdictions. Here, we present the RuralAI architecture deployment for tomato crop monitoring, featuring sensor field units for soil, crop, and weather data collection. HFL with personalization is used to offer localized and adaptive insights. Model management, aggregation, and transfers are facilitated via a flexible approach, enabling seamless communication between local devices, edge nodes, and the cloud.

60 APPLIED LIFE SCIENCES↗

Global-View Coefficients: A Data Management Solution for Parallel Quantum Monte Carlo Applications

Quantum Monte Carlo (QMC) applications perform simulation with respect to an initial state of the quantum mechanical system, which is often captured by using a cubic B-spline basis. This representation is stored as a read-only table of coefficients, and accesses to the table are generated at random as part of the Monte Carlo simulation. Current QMC applications, such as QWalk and QMCPACK, replicate this table at every process or node, which limits scalability because increasing the number of processors does not enable larger systems to be run. We present a partitioned global address space (PGAS) approach to transparently managing this data using Global Arrays in a manner that allows the memory of multiple nodes to be aggregated. We develop an automated data management system that significantly reduces communication overheads, enabling new capabilities for QMC codes. Experimental results with QWalk and QMCPACK demonstrate the effectiveness of the data management system.

Niu, Qingpeng↗