SEARCH · Engineering Papers

Results for “datacenter”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

29 records · Page 2

All sources NASA NTRS DOE OSTI

DOE OSTI2025

Congestion Management Solutions for Enhanced Distribution System Operations with Aggregated Distribution Grid Resources Providing Grid Services and Market Participation

Microgrids and other aggregations of distribution grid resources (DGRs) are poised to actively participate in electricity markets and provide essential grid services in the coming years. In fact, DGRs already play such a role through behind-the-meter (BTM) demand response programs and small-scale BTM dispatchable generation initiatives. At the same time, the rapid growth of artificial intelligence (AI) and cryptocurrency datacenters imposes significant, often unpredictable, demands on the power distribution system. Aggregated DGRs can serve as flexible resources that help mitigate these pressures by using available transmission and distribution capacity more efficiently, supporting resource adequacy and other reliability services, and providing bridge strategies while long-term transmission infrastructure is being developed. The impacts this activity will have on distribution networks are not fully understood and could present significant challenges for distribution utilities due to capacity constraints and the need for congestion management. Technical issues include reverse power flow, variability and possible degradation of equipment integrity, voltage violations, and customer power quality concerns. These issues will likely intensify as electricity market operators across the United States implement Federal Energy Regulatory Commission Order 2222 over the next few years.

24 POWER TRANSMISSION AND DISTRIBUTION↗

NASA NTRS2018

Technologies and Methods Used at the Laboratory for Atmospheric and Space Physics (LASP) to Serve Solar Irradiance Data

The Laboratory for Atmospheric and Space Physics (LASP) at the University of Colorado in Boulder, USA operates the Solar Radiation and Climate Experiment (SORCE) NASA mission, as well as several other NASA spacecraft and instruments. Dozens of Solar Irradiance data sets are produced, managed, and disseminated to the science community. Data are made freely available to the scientific immediately after they are produced using a variety of data access interfaces, including the LASP Interactive Solar Irradiance Datacenter (LISIRD), which provides centralized access to a variety of solar irradiance data sets using both interactive and scriptable/programmatic methods. This poster highlights the key technological elements used for the NASA SORCE mission ground system to produce, manage, and disseminate data to the scientific community and facilitate long-term data stewardship. The poster presentation will convey designs, technological elements, practices and procedures, and software management processes used for SORCE and their relationship to data quality and data management standards, interoperability, NASA data policy, and community expectations.

Pankratz, Chris↗

NASA NTRS2021

NASA SpaceCube Edge TPU SmallSat Card for Autonomous Operations and Onboard Science-Data Analysis

Using state-of-the-art artificial intelligence (AI)frameworks onboard spacecraft is challenging because common spacecraft processors cannot provide comparable performance to datacenters with server-grade CPUs and GPUs available for terrestrial applications and advanced deep-learning networks. This limitation makes small, lo w-p o we r AI microchip architectures, such as the Google Coral Edge Tensor Processing Unit (TPU), attractive for space missions where the application-specific design enables both high-performance and power-efficient computing for AI applications. To address these challenging considerations for space deployment, this research introduces the design and capabilities of a CubeSat-sized Edge TPU-based co-processor card, known as the SpaceCube Low-power Ed g e Artificial Intelligence Resilient Node (SC-LEARN). This design conforms to NASA’s CubeSat Card Specification (CS2) for integration into next-generation SmallSat and CubeSat systems. This paper describes the overarching architecture and design of the SC-LEARN, as well as, the supporting test card designed for rapid prototyping and evaluation. The SC-LEARN was developed with three operational modes: (1) a high-performance parallel-processing mode,(2)a fault-tolerant mode for onboard resilience, and (3) a power-saving mode with cold spares. Importantly, this research also elaborates on both training and quantization of Tensor Flow models for the SC-LEARN for use onboard with representative, open-source datasets. Lastly, we describe future research plans, including radiation-beam testing and flight demonstration.

Advanced avionics↗

NASA NTRSDate not supplied

A Framework for Assessing Earth Observation Metadata Quality: Implications for Data Discovery and Open Science

The Common Metadata Repository (CMR) contains metadata records describing NASA’s collection of over 8,000 Earth observation data products. The Analysis and Review of CMR (ARC) Team at Marshall Space Flight Center assesses the quality of these metadata records. Metadata, rather than the data itself, is indexed for search in both discipline-specific datacenters and global or aggregated catalogs (such as Earth data Search), making it essential for determining whether a data product is appropriate for a given research question or application need. Since metadata connects users to data, it should be as accurate and complete as possible in addition to meeting minimum database requirements. The ARC team has developed a metadata quality framework by which to assess quality. The framework consists of a set of quality criteria that converge around the dimensions of correctness, completeness, and consistency, with the goal of improving the discoverability, accessibility, and usability of NASA’s Earth Observation data. The application of the framework has resulted in a measurable improvement in NASA’s metadata quality. Key aspects of the framework’s success are the ability to systematically evaluate metadata and provide actionable quality improvement recommendations. Lessons learned from the project will be shared along with implementation details which may be relevant to other science disciplines. By aiming to make data more discoverable and accessible to a broad user community, the ARC metadata quality framework helps contribute to NASA’s commitment to open science.

Jeanne Le Roux↗

NASA NTRSDate not supplied

Fostering Better Collaboration in Software Development Cycles Between Scientists and Programmers to Ensure the Integrity of and Promote the Development of New Scientific Data Products.

Misaligned incentives lead to reduced interaction between scientists and programmers on modern NASA science data-product development teams. Typically, situations arise where the scientist is not incentivized to learn modern coding practices and the programmer does not understand the science algorithms in the code. A programmer is responsible for the deliverable code thus setting a tradeoff between the desire for code improvement versus fear of compromising the integrity of data-product while the scientist continues to rely on their legacy codebases owing to the complexity of using the delivered code outside the processing environment and lack of validation modules. The NASA/CERES-TISA project has adopted a collaborative approach, with scientists and programmers both utilizing the same software repository with multiple branches, some optimized for delivery to a processing datacenter and others for scientific product development and validation. A team of scientists and programmers jointly review any new science code updates for integration into the codebase and strive to improve practices through promoting algorithm understanding, better institutional knowledge exchange and documentation, modularization, and developing data processing flow-dictated validation and debugging methods. This leads to a reduction in the personnel single point failures and reduced development time for creation of new science data-products.

CERES↗

DOE OSTI2025

Job Scheduler-Driven Power Gateway for High Performance Computing

Power gateways in the form of a microgrid can incorporate multiple distributed energy resources (DER) in either grid forming or grid following mode and support high performance computing (HPC) power profiles including the large load-follow requirements observed in multi-user HPC systems. The microgrid’s flexibility to operate in either grid forming or grid following mode and to actively switch between these modes enables baseline power from multiple non-baseline DER while maintaining high power quality metrics for the HPC system. But this enormous flexibility in demand response and time of use shifting is generally programmed independently of any integration with an HPC job scheduler which can better inform the load shaping by the microgrid. While there are many existing approaches where the HPC job scheduler takes in information from the grid to make queue scheduling decisions, this work takes the opposite view and explores a scheduler where the jobs in the queue can directly impact the settings of the grid. Several HPC scheduler strategies are tested where the jobs in the queue directly impact the settings of a microgrid designed for HPC operation which is driving a datacenter with three classes of HPC architectures. The scheduler operation is shown using a microgrid with 64 kW of solar capacity and 320 kWh of battery over a period of 21 days operating with significant low-follow swings, a throttled grid, cloudy conditions, switching between grid following and grid forming modes, and a wide range of battery states-of-charge all while maintaining high quality power metrics. The scheduler provides a mechanism for the job queue to directly impact a power gateway like a microgrid and to improve HPC power outcomes such as maximizing renewable energy usage

microgrid↗

DOE OSTI2024

Microgrid Integration with High Performance Computing Systems for Microreactor Operation

Multiple nuclear microreactor concepts are currently being developed across several sizes and fuel types with high performance computing (HPC) systems anticipated to be end-users of the power. Nuclear microreactors are small in size, portable, produce less than 10 MW electric, operate autonomously, and have a refueling interval of as many as 10 years. However, their load-follow is also generally limited to 10%/minute or worse whereas the power variance in HPC systems easily exceeds this constraint under normal operations. This study explores an approach that requires no load-follow from the microreactor but integrates the HPC system with a microgrid built from commercial-off-the-shelf components. Three typical HPC architectures are explored in the context of microgrid operation in this study. Components of power quality and transient response are empirically measured for five different HPC load-follow response levels using a self-contained mobile datacenter connected to the microgrid capable of integration with a nuclear microreactor.

microgrids↗

DOE OSTI2026

Electromagnetic Transient Modeling of Large Data Centers for Grid-Level Studies

The magnitude and complexity of electricity usage patterns from large data centers are having significant impacts on the operation and dynamics of the power grid; grid operators and planners require a range of specialized data center models to properly evaluate these impacts and specify technical solutions as needed. Towards addressing this need, Pacific Northwest National Laboratory (PNNL) has developed a library of electromagnetic transient (EMT) models for grid-level studies of data centers called the data center model library (DML). This report describes how the DML was created and how it may properly be used. The models present in the DML are generic models; subject matter expertise and additional technical data are needed to modify these models before they can represent any real data center. However, they will significantly reduce the level of effort required to develop site-specific models and can serve as a common starting point to guide industry towards a more refined consensus. Most of the models within DML are dedicated to representing the power electronics interfaces commonly used in modern data centers, such as double-conversion uninterruptible power supplies and single-phase power factor correction converters. These models are intended for use in grid-level studies and are a simplified aggregation of many small components. That said, background material on the physical and electrical design of large data centers is provided as companion material so that users can be aware of many of the details which have been omitted or streamlined as a matter of practical necessity. Additionally, guidance on the application of EMT analysis for data center interconnection studies is provided, which aids users in identifying when the DML is necessary and what sort of additional model development may be necessary for conducting real-world studies.

24 POWER TRANSMISSION AND DISTRIBUTION↗

DOE OSTI2026

Electromagnetic Transient Modeling of Large Data Centers for Grid-Level Studies: Beta Release

The magnitude and complexity of electricity usage patterns from large data centers are having significant impacts on the operation and dynamics of the power grid; grid operators and planners require a range of specialized data center models to properly evaluate these impacts and specify technical solutions as needed. Towards addressing this need, Pacific Northwest National Laboratory (PNNL) has developed a library of electromagnetic transient (EMT) models for grid-level studies of data centers called the data center model library (DML). This report describes how the DML was created and how it may properly be used. This report details the DML’s beta release, completed in July 2026. This is a revision and expansion of the alpha release, which was made available in January 2026 The models present in the DML are generic models; subject matter expertise and additional technical data are needed to modify these models before they can represent any real data center. However, they will significantly reduce the level of effort required to develop site-specific models and can serve as a common starting point to guide industry towards a more refined consensus. Most of the models within DML are dedicated to representing the power electronics interfaces commonly used in modern data centers, such as double-conversion uninterruptible power supplies and single-phase power factor correction converters. These models are intended for use in grid-level studies and are a simplified aggregation of many small components. That said, background material on the physical and electrical design of large data centers is provided as companion material so that users can be aware of many of the details which have been omitted or streamlined as a matter of practical necessity. Additionally, guidance on the application of EMT analysis for data center interconnection studies is provided, which aids users in identifying when the DML is necessary and what sort of additional model development may be necessary for conducting real-world studies.

electromagnetic transients↗

NASA NTRS2012

Cloud Computing for Mission Design and Operations

The space mission design and operations community already recognizes the value of cloud computing and virtualization. However, natural and valid concerns, like security, privacy, up-time, and vendor lock-in, have prevented a more widespread and expedited adoption into official workflows. In the interest of alleviating these concerns, we propose a series of guidelines for internally deploying a resource-oriented hub of data and algorithms. These guidelines provide a roadmap for implementing an architecture inspired in the cloud computing model: associative, elastic, semantical, interconnected, and adaptive. The architecture can be summarized as exposing data and algorithms as resource-oriented Web services, coordinated via messaging, and running on virtual machines; it is simple, and based on widely adopted standards, protocols, and tools. The architecture may help reduce common sources of complexity intrinsic to data-driven, collaborative interactions and, most importantly, it may provide the means for teams and agencies to evaluate the cloud computing model in their specific context, with minimal infrastructure changes, and before committing to a specific cloud services provider.

cloud computing↗

DOE OSTI2026

Data Center High-Temperature Liquid Cooling and Heat Reuse Techno-Economic Study: Preprint

Data centers are energy-intensive facilities with growing demands for efficiency and cost-effective operations. Smaller, more distributed edge inference data centers are expected to proliferate as AI applications require low latency closer to the user of AI tools, which presents a growing opportunity to explore the systems implications of liquid cooling on water and energy use. This study analyzes the implementation of high-temperature liquid cooling systems in a prototypical inference 1-MW data center and explores the potential for heat reuse across varying climates with a goal to optimize energy efficiency, reduce capital and operational costs, and identify opportunities for high-performance cooling and water use reduction infrastructure. This analysis evaluated configurations utilizing a peak day hourly sizing and systems performance spreadsheet to evaluate design and operational conditions from which component sizes, installed cost, operational cost, and performance metrics were determined for the Base case and the Elevated case. The techno-economic analysis included heat reuse applications across a range of heat recovery temperatures and heat rejection options. The analysis shows that high-temperature liquid cooling allows for improved energy efficiency, lower water consumption, and lower capital costs compared to traditional cooling approaches. Transitioning to elevated water inlet/outlet temperatures (50 degrees C/60 degrees C) eliminates the need for chillers, cooling towers, and heat recovery equipment in many scenarios across three distinct climate zones. This results in up to 75% capital cost savings for the cooling and heat recovery equipment, and with significantly reduced water consumption, especially in non-heat reuse applications. Heat generated from data centers can also be repurposed for space heating, domestic hot water, and other applications, and is most cost-effective when data center outlet temperatures exceed 55-60 degrees C.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗