Search NASA⌕ Search

SEARCH · Search NASA

Results for “building operations optimization”

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 433 records · Page 24
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Speeding up particle track reconstruction using a parallel Kalman filter algorithm

One of the most computationally difficult problems expected for the High-Luminosity Large Hadron Collider (HL-LHC) is determining the trajectory of charged particles during event reconstruction. Algorithms used at the LHC today rely on Kalman filtering, which builds physical trajectories incrementally while incorporating material effects and error estimation. Recognizing the need for faster computational throughput, we have adapted Kalman-filter-based methods for highly parallel, many-core SIMD architectures that are now prevalent in high-performance hardware. In this paper, we discuss the design and performance of the improved tracking algorithm, referred to as mkFit. A key piece of the algorithm is the Matriplex library, containing dedicated code to optimally vectorize operations on small matrices. The physics performance of the mkFit algorithm is comparable to the nominal CMS tracking algorithm when reconstructing tracks from simulated proton-proton collisions within the CMS detector. We study the scaling of the algorithm as a function of the parallel resources utilized and find large speedups both from vectorization and multi-threading. mkFit achieves a speedup of a factor of 6 compared to the nominal algorithm when run in a single-threaded application within the CMS software framework.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2021

Transportation Hub Infrastructure Expansion: Decision Support Under Uncertainty

The Athena project (www.athena-mobility.org) has worked to investigate the relationship between the Dallas-Fort Worth Airport (DFW) and the greater Dallas area in order to better understand and therefore better inform future decision-making regarding the critical infrastructure that influence mobility between the airport and the city. Through this work, infrastructure related to curbside pickup and drop-off, parking, public transit, and the road network congestion were identified as critical to the operation of the DFW transportation hub. The infrastructure analysis and expansion aspect of the Athena project is focused on the restructuring of the CTA curb as a hierarchical curb and the building or repurposing of parking infrastructure as the interplay between these two areas. Many sources of uncertainty exist that may impact future airport and transportation hub operations, such as passenger volume growth, population demographic changes over time, electric vehicle (EV) adoption rates, and autonomous vehicle (AV) adoption rates. Due to these sources of uncertainty, we have selected for our research a modeling framework that can capture various types of uncertainty and hedge against those uncertainties in the optimization process. We analyze road network and curb congestion, the rise of transportation networking companies, trends in parking usage, existing policies around this infrastructure, airport revenue streams, and other contributing factors to enable infrastructure decision making with less uncertainty. To accomplish this wholistic analysis, we have developed a novel multi-stage, multi-period stochastic optimization model which considers the airport's decisions from 2025-2045 under different possible future macro trajectories and day-to-day variations in operational conditions captured as "annual representation of operations" scenarios with respective probabilities. This model has also been designed to leverage the outputs of various efforts under the Athena project to create a combined decision framework for infrastructure decisions. These various efforts include the route optimization model, the ASPIRES simulation, the mode choice model, and the SUMO traffic simulation. Our computational experiments of this system at scale have resulted in a working version of our infrastructure model which enables the explicit representation and consideration of various sources of uncertainty in the decision process to enable robust, flexible decision-making. This model has been effectively run on NREL's HPC system, Eagle, with large numbers of stochastic scenarios and shows promise as a scalable tool for robust consideration of uncertainties in airport planning. We have tested our model using 30,240 operational circumstances in total, resulting in a problem with more 200 million variables. This model was solved in several different configurations, and a workflow to simulate the performance of the infrastructure model results was developed and deployed. In general, our results indicate that a combination of remote parking, remote curb infrastructure, and dynamic pricing can generate revenue, reduce emissions, accommodate emerging technologies such as AVs and EVs, and manage airport passenger growth over time. We note the success of the proposed strategy depends on the data collection and forecasting abilities of DFW. We have also seen that the AV adoption by TNCs might necessitate larger amounts of remote curb. The results of this work inform strategies for airport infrastructure decision making, as well as demonstrate the value of an adaptable model, but also indicate that there are avenues remaining where further research would be of value.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

Optimizing Control for Efficient Load Shifting with Thermal Energy Storage in Existing HVAC Systems

This project developed the integration of a direct-contact heat exchanger (DCHX) based thermal energy storage (TES) system with a chiller–air handling unit (AHU) plant to evaluate its potential for reducing building energy costs. Detailed physical models of the TES unit, building envelope, and HVAC components were developed alongside simplified control-oriented models to support both high-fidelity simulation and real-time optimization. Two control strategies were implemented and compared: a rule-based control (RBC) aligned with utility time-of-use (ToU) rates, and a model predictive control (MPC) framework leveraging forecasts of building load, weather, and internal gains. Simulation results show that the RBC strategy reduced daily electricity costs by around 30% by shifting cooling production from on-peak to off-peak hours. In contrast, the MPC strategy achieved significantly greater performance, reducing daily operating costs by up to 44% and peak-hour costs by more than 60%. Both strategies maintained indoor thermal comfort within acceptable limits, with MPC further improving load distribution and reducing equipment cycling. The outcomes confirm that TES integration, particularly when coordinated with advanced predictive control, can provide substantial cost savings and on-peak demand reduction. . These findings directly support the U.S. Department of Energy’s goals for grid-interactive efficient buildings and demonstrate the potential of TES-enabled HVAC systems for scalable deployment across the commercial building.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
NASA NTRSDate not supplied

Spacecraft Cabin Ventilation Fan Research at NASA

NASA has recently made the geometry and solid model for a spacecraft cabin ventilation fan prototype available to the public via electronic file downloads from the NASA Technical Report Server. This fan can be used for research and development by many organizations. The fan is 3.5 inches in diameter, 9 inches long, and weighed 3.6 lb. The NASA Quiet Space Fan was designed to make 3.64 inches of water pressure rise at 150.3 cfm of airflow at 12,000 rpm at standard air conditions of 70ºF and 14.7 psia. The performance of the metal version of the fan was measured to be 3.48 inches of water at 150.6 cfm at design speed. A low-noise blade-vane count was chosen to try to reduce tone noise generated by this fan by cutting off the first three blade passing frequency tones. In-duct microphone array measurements indicated that the most evident tones occur for frequencies of 1800 Hz (1 BPF) and 7200 Hz (4 BPF). Using reverberant room standard testing methods, the A-weighted sound power level for the fan operating at design point conditions was measured to be 71 dBA. This report describes the current set of publicly available information for this fan. The fan was designed, optimized, and tested with tools and techniques that NASA has traditionally used for turbofan engine research. This is one way that technology developed for aerospace applications can be used more broadly, since quiet and efficient fans are needed for many ventilation systems on spacecraft, aircraft, watercraft, land vehicles, and buildings.

fan, noise, ventilation↗
NASA NTRSDate not supplied

A Cost and Benefit Analysis of Orbital Debris Remediation, Mitigation,Tracking, and Characterization

Orbital debris may collide with crewed and robotic spacecraft, placing them at risk. The wide range of debris, from 9,000-kilogram rocket bodies to millions of millimeter-size debris, has led to a similarly wide range of proposed actions for addressing the risks posed by debris. However, the costs and benefits of these actions have historically been unknown. This is a challenge for decision makers who are choosing which actions to support through technology development or policy changes. NASA’s Office of Technology, Policy, and Strategy is addressing these technical and economic uncertainties by building a capability to (1) complete rigorous calculations of the net present value of each action, (2) identify an optimal portfolio of actions to reduce risk, and (3) quantitatively analyze policies related to space sustainability. This report describes our progress toward that capability and to solicit feedback from the space and economic communities. Our previous work(Colvin, Karcz, and Wusk. 2023), referred to here as Phase 1, assessed the costs and benefits of performing debris remediation on operationally relevant timescales. The current paper summarizes the work of Locke and Colvin (2024), which contains major updates to the risk model used in Phase 1 and expands the breadth of actions considered to include mitigating the creation of debris, improving the ability to track debris, and more methods for cleaning up existing debris. We demonstrate that our approach of measuring risks in dollars allows for the effectiveness of seemingly incommensurate actions to be compared and generates insights that other approaches to measuring risk have missed.

Orbital Debris↗
DOE OSTI2023

Analysis of Building Model Forecasts using Autonomous HVAC Optimization System for Residential Neighborhood

Heating, ventilation, and air conditioning (HVAC) systems account for the highest share of home energy consumption in the United States. Optimized HVAC control can provide thermal improved comfort to the occupants, improve energy efficiency, reduce energy cost, and support grid services. In this paper, we discuss a multi-agent and cloud-based software framework that has been deployed in occupied residential neighborhood. This system enables automatic data collection, learning, optimization, and dispatches signals to neighborhood devices. HVAC optimization is based on model predictive control (MPC). Since the operational performance of MPC depends on model forecasting accuracy, it is crucial to evaluate the model continuously and modify or retrain it as necessary. In this research, we developed an automated workflow to evaluate the performance of temperature and power forecasts based on measured data in the real world. This will provide researchers with a deeper understanding of the model and how it can be improved.

Lebakula, Viswadeep↗
DOE OSTI2022

A Novel Steam Condenser with Loop Thermosyphons and Film-Forming Agents for Improved Heat Transfer Efficiency and Durability

In this DOE/NETL BAA program, Advanced Cooling Technologies, Inc. (ACT), with Suez Water Technologies & Solutions (Suez), developed film-forming amine coatings applied to condenser surfaces that incorporated loop thermosyphons to enhance steam surface condenser performance and efficiency for coal-fired power plants. The condensation phenomenon resulting from these coatings promotes efficient dropwise condensation without additional thermal resistance through the use of film-forming amines (FFA). The loop thermosyphon technology was being developed to supplant current pumped cooling water systems to reduce operations and maintenance issues, reduce energy use, and promote high thermal performance. Both of ACT’s advanced coating and loop thermosyphon technologies were developed for application to wet cooling and indirect dry cooling systems. Internal steam condensation and subsequent heat dissipation complete the thermodynamic power cycle used by power plant cooling systems. The successful 3-year project period delved into understanding the experimental performance gain using film-forming amines applied to flat and tubular condenser surfaces under realistic operating conditions. The performance gain was used to create a technoeconomic analysis to more clearly understand the market value for power plants using FFA coatings to achieve dropwise condensation on condenser surfaces. A tall (50 ft.) loop thermosyphon was designed, fabricated, and tested to optimize the performance characteristics by tuning the working fluid charge and determining the temperature drop across the loop at different heat loads. Commercialization opportunities were explored to transition the loop thermosyphon work into the building HVAC market.

20 FOSSIL-FUELED POWER PLANTS↗
DOE OSTI2023

IDAES-PSE 2.2.0 Release

The Institute for the Design of Advanced Energy Systems (IDAES) Integrated Platform is a versatile computational environment offering extensive process systems engineering (PSE) capabilities for optimizing the design and operation of complex, interacting technologies and systems. IDAES enables users to efficiently search vast, complex design spaces to discover the lowest cost, most environmentally sustainable solutions while supporting the full process modeling lifecycle, from conceptual design to dynamic optimization and control. The extensible, open platform empowers users to create models of novel processes and rapidly develop custom analyses, workflows, and end-user applications. New features in the 2.2 release include: DiagnosticsToolbox DiagnosticsToolbox (idaes.core.util.model_diagnostics. DiagnosticsToolbox is a new self-guiding toolbox for assisting with diagnosing modeling issues. The toolbox contains a wide range of tools for identifying common modeling issues and conveying these to the user, and more features will be added in the coming months. To try the new toolbox, use the following steps: from idaes.core.util import DiagnosticsToolbox Build an instance of your model Ensure your models has zero degrees of freedom (as best you can) dt = DiagnosticsToolbox(model) dt.report_structural_issues() Follow the prompts Deprecations DegeneracyHunter is being deprecated in favor of the new DiagnosticsToolbox. The features and capabilities of DegeneracyHunter will be moved over to the DiagnositcsToolbox in the coming months. Removal of usage of Pyomo's expr.current due to deprecation upstream. Removed support for radial basis functions in ALAMOpy as these did not work in the current implementation. Those wishing to use radial basis functions should use Pysmo instead. New Capabilities MSContactor model for unit operations involving mass transfer between 2 or more streams, such as membrane, solvent extraction and solids leaching operations. Other Improvements Improved formulation for solubility products with additional parameters for better scaling of terms Better backward compatibility, verification and robustness testing of core models and APIs Added support for Separator models without energy balances

DiagnosticsToolbox↗
DOE OSTI2022

A case study of high-temperature polyetherimide film capacitor fabrication

Capacitors, as the basic building block of power electronics and electrical systems, are the major constraint of the increasingly integrated power systems that require new capable polymer dielectric films operating at higher temperatures >125 °C. Despite tremendous research efforts on the lab scale, there remains a considerable barrier and knowledge basis for converting polymeric films to functioning capacitors desired in actual applications. Here, the authors tackled the polyetherimide (PEI) film scale up issues and developed various engineering processes for film de-wrinkling, optimal metallization, static elimination, and capacitor fabrication improvement. The authors discovered various fabrication challenges such as the electrostatic charge, metallization scheme, winding tension, and end electrode disconnects. To fabricate a high yield of PEI capacitor bobbins, it is necessary to utilize static eliminators (radioactive ionizers), moderately thick aluminum metallization (15–30 Ω/sq), and winding tension (60–100 g). It is also effective to evaluate capacitance, dielectric loss, equivalent series resistance, and thermal cycling stability of capacitors. This work sheds some light on converting PEI films and other polymeric films to capacitors on a scale up fabrication effort. The practical learning from film handling to capacitor fabrication in this work provided the necessary knowledge for manufacturing high-temperature polar film capacitors.

36 MATERIALS SCIENCE↗
NASA NTRS2013

Scheduling Operations for Massive Heterogeneous Clusters

High-performance computing (HPC) programming has become increasingly difficult with the advent of hybrid supercomputers consisting of multicore CPUs and accelerator boards such as the GPU. Manual tuning of software to achieve high performance on this type of machine has been performed by programmers. This is needlessly difficult and prone to being invalidated by new hardware, new software, or changes in the underlying code. A system was developed for task-based representation of programs, which when coupled with a scheduler and runtime system, allows for many benefits, including higher performance and utilization of computational resources, easier programming and porting, and adaptations of code during runtime. The system consists of a method of representing computer algorithms as a series of data-dependent tasks. The series forms a graph, which can be scheduled for execution on many nodes of a supercomputer efficiently by a computer algorithm. The schedule is executed by a dispatch component, which is tailored to understand all of the hardware types that may be available within the system. The scheduler is informed by a cluster mapping tool, which generates a topology of available resources and their strengths and communication costs. Software is decoupled from its hardware, which aids in porting to future architectures. A computer algorithm schedules all operations, which for systems of high complexity (i.e., most NASA codes), cannot be performed optimally by a human. The system aids in reducing repetitive code, such as communication code, and aids in the reduction of redundant code across projects. It adds new features to code automatically, such as recovering from a lost node or the ability to modify the code while running. In this project, the innovators at the time of this reporting intend to develop two distinct technologies that build upon each other and both of which serve as building blocks for more efficient HPC usage. First is the scheduling and dynamic execution framework, and the second is scalable linear algebra libraries that are built directly on the former.

Humphrey, John↗
DOE OSTI2025

Microreactor Optimization Using Simulation And Economics (mouse)

Microreactor Optimization Using Simulation and Economics (MOUSE) is a tool that integrates both nuclear microreactor design and reactor economics to provide comprehensive evaluations and optimizations. This tool enables stakeholders to explore the interplay between technical and economic variables, guiding them towards effective and competitive microreactor solutions. For the reactor core simulations, MOUSE leverages the OpenMC Monte Carlo Particle Transport Code to perform detailed core simulations for various microreactor designs. The included OpenMC models are 2D core designs of a Liquid Metal Thermal Microreactor (LMTR), a Gas-Cooled TRISO-Fueled Microreactor (GCMR), and a Heat Pipe Microreactor. Beyond core design, MOUSE includes simplified calculations for: - Calculating the masses of heat exchangers within the system. - Mechanical power of pumps. - Estimating the area occupied by various buildings within the nuclear plant. For the economic analysis, MOUSE provides detailed bottom-up cost estimates, encompassing a wide range of costs including preconstruction costs, direct costs, indirect costs, training costs, financial costs, operation & maintenance (O&M) costs, and fuel costs. These cost estimations are developed using data from the MARVEL project and additional literature sources, enabling the calculation of total capital costs and levelized cost of energy for both first-of-a-kind and nth-of-a-kind microreactors. MOUSE also enables analysis of the cost drivers and competitiveness in the electricity market. MOUSE allows users to modify a wide array of technical and economic parameters to evaluate different scenarios and their impacts. Examples of these parameters include: Fuels, coolants, or reflector materials Enrichment levels Control drum materials and geometry Fuel pin geometry and materials Moderator pin geometry and materials Reactor core and reflector dimensions Packing factor for the TRISO particles Nuclear reactor power and reactor burnup Number of sensors Shielding thickness Reactor vessel and guard vessel dimensions Operational staff requirements Number of emergency shutdowns Levelization period Interest rate Construction duration Since MOUSE is powered by the WATTS toolkit, it supports optimization studies, parametric analyses, and uncertainty calculations/propagation. The optimization techniques enable users to identify optimal design and economic configurations. The parametric analysis tools allow users to explore the sensitivity of various parameters, while uncertainty propagation helps quantify the impact of uncertainties on overall performance and cost. User Interface and Workflow: Currently, MOUSE is a command-line-based tool. Users can input various reactor design or economic parameters, modify the designs, run simulations, and visualize results through comprehensive data visualization and reporting capabilities. The typical workflow involves setting up the reactor model, defining economic parameters, running simulations, and analyzing the results to make informed decisions. By combining advanced design calculations with detailed economic modeling, MOUSE provides a robust framework for optimizing nuclear microreactor technologies, enhancing their competitiveness, and guiding stakeholders towards innovative and cost-effective solutions.

Hanna, Botros [Idaho National Laboratory (INL), Id↗
DOE OSTI2022

Development and assessment of a reactor system prognosis model with physics-guided machine learning

Autonomous control systems provide recommendations to help operators in decision-making during plant operations ranging from normal operation to accident management. An important step of autonomous control is prognosis. In nuclear engineering domain, prognosis is the process of predicting future conditions of a system or equipment based on present signs and symptoms of a fault. The prognosis model allows predicting future reactor states for possible candidate control strategies so that the outcomes can be evaluated to determine the best control strategy. The prognosis model requires representing direct relationships between the symptoms and the predictions. In nuclear engineering, computational simulations are approximate representations of the operation of the real system. However, prognosis with computational simulations requires high computation power and time due to possible large number of scenarios. Necessary computation resources can be reduced with machine learning (ML) approach for fast predictions by building a surrogate function using the simulation data. A critical issue is, ML models are ignorant of physical knowledge, and these models approximate statistical relationships between the system variables. This ignorance can produce results that are inconsistent with physical laws, even if an optimal result is achieved from a mathematical point of view. Physics-guided machine learning (PGML) is an approach to tackle this issue. Here, this work formulates and illustrates a framework to guide development and assessment of the ML-based prognosis model for autonomous control systems. The development of the prognosis model considers the training of a ML model which consists of optimizing many aspects of the ML approach. The assessment of the prognosis model considers training data limitations and uncertainties of the ML approach. Prognosis models with standalone ML and PGML are developed and assessed on the loss-of-flow scenario of Experimental Breeder Reactor II. The results indicate that PGML based prognosis model has the best performance compared to other prognosis models.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2021

Electrical Measurement and Verification of Energy in DC Buildings

Today's selection of DC buildings features a diverse set of electrical topologies and turnkey solutions, and each has specific design trade-offs and optimizations. Designers desperately need standardized metrics and procedures for measurement and verification (M&V) to analyze and compare the advantages of each DC solution to traditional AC building networks. This work develops the Measurement-Informed Modeling (MIM) method, which can be used to determine full-building efficiency and energy savings. The MIM M&V procedure develops a building model, and refines the model with metered data. This work demonstrates the MIM method by measuring the full-building efficiency of two DC buildings operated by the Institute of Building Research in Shenzhen, China. The MIM procedure can ultimately be used to compare and improve the efficiency of various DC topologies.

buildings↗
DOE OSTI2025

Helium recovery system at IB3A

The growing demand for sustainable cryogenic operations at Fermilab has underscored the need to improve helium management, particularly at the Industrial Building 3A (IB3A) test facility. IB3A characterizes and tests superconductors, cables, and coils for projects such as the HL-LHC AUP and Mu2e, yet currently relies on 500 L Dewars whose boil-off is vented to atmosphere, wasting a critical, non-renewable resource and increasing the cost of testing. A project is therefore under way to link IB3A to an existing purification and liquefaction station in a neighboring building through a dedicated pipeline. Captured helium will be transferred, purified, reliquefied, and returned for reuse, cutting losses and operating costs. This paper details the first two project phases: "Design and Engineering" and "Procurement and Installation." The design phase finalizes pipeline specifications, establishes flow-control requirements, and resolves integration challenges with existing cryogenic infrastructure. The procurement and installation phase covers material sourcing, pipeline construction, and deployment of control and monitoring systems to assure reliable, efficient operation. Key technical hurdles-route optimization, pressure drop mitigation, and interface compatibility-are discussed alongside implemented solutions. Implementing the pipeline and upgrading IB3A will dramatically reduce helium consumption and therefore testing cost, strengthening Fermilab's capacity to support frontier science far into the future.

Porwisiak, D. [Fermilab; Wroclaw Tech. U.]↗
DOE OSTI2022

Coupling subsurface and above-surface models for optimizing the design of borefields and district heating and cooling systems in the presence of varying water-table depth. In: Proceedings, 46th Workshop on Geothermal Reservoir Engineering

Dynamic energy simulation is important for the design and sizing of district heating and cooling systems with geothermal heat exchange. Current modeling approaches in building and district energy simulation tools typically consider heat conduction through the ground between boreholes, without flow of groundwater. On the other hand, detailed simulation tools for subsurface heat and mass transfer exist, but these fall short in simulating above-surface energy systems. To support the design and operation of such systems, we have developed a coupled model including a software package for building and district energy simulation, and software for detailed heat and mass transfer in the ground. For the first, we use the open-source Modelica Buildings Library, which includes dynamic simulation models for building and district energy and control systems. For the heat and mass transfer in the soil, we use the TOUGH simulator. TOUGH can model heat and multi-phase, multi-component mass transport for a variety of fluid systems, as well as chemical reactions, in fractured porous media. In previous work, we described the coupling of these software packages, including how time-dependent boundary conditions for the borehole walls are synchronized for use in Modelica and TOUGH. We verified that the coupled Modelica/TOUGH code produced consistent results with the original Modelica code for an idealized problem in which heat transfer was purely by conduction in a uniform geologic medium. Here, we examine less idealized problems for which TOUGH’s advanced capabilities for modeling fluid flow are required. The first problem has a shallow vadose zone, and the second problem has a thicker vadose zone with a water-table depth that varies in time, which requires a fine vertical grid discretization for the TOUGH model.

Doughtry, Christine↗
DOE OSTI2020

Hardware Design and Demonstration of a 100kW, 99% Efficiency Dual Active Half Bridge Converter Based on 1700V SiC Power MOSFET

High efficiency, high density and galvanically isolated power converters are attractive for numerous medium voltage high power applications. This paper presents a 1.5kVdc, 100kW bidirectional Dual Active Half Bridge (DAHB) using newly developed 1.7kV SiC MOSFET modules. The DAHB achieved an efficiency of 98.6% at full load of 100kW and a maximum of 99% at 45kW, operating as an isolated DC/DC converter. The converter also achieved an efficiency of 97.8% operating as an isolated DC/AC inverter. An optimized PCB-based busbar design has significantly reduced the voltage overshoot across the device, making the design suitable for 1500Vdc input application such as 1500V PV inverters. Typical partial discharge inception voltage (PDIV) of the optimized PCB busbar is 1.7kVpeak with total charge <10pC. The power density of the DAHB converter is 1.8MVA/m 3 which is much higher than traditional two-stage industry products. The DAHB converter can be used as a building block for a 4.16kV/1MVA utility scale PV systems with input parallel and output series configuration.

Xu, Wei↗
DOE OSTI2023

Unified Communication Optimization Strategies for Sparse Triangular Solver on CPU and GPU Clusters

This paper presents a unified communication optimization framework for sparse triangular solve (SpTRSV) algorithms on CPU and GPU clusters. The framework builds upon a 3D communication-avoiding (CA) layout of Px × Py × Pz processes that divides a sparse matrix into Pz submatrices, each handled by a Px × Py 2D grid with block-cyclic distribution. We propose three communication optimization strategies: First, a new 3D SpTRSV algorithm is developed, which trades the inter-grid communication and synchronization with replicated computation. This design requires only one inter-grid synchronization, and the inter-grid communication is efficiently implemented with sparse allreduce operations. Second, broadcast and reduction communication trees are used to reduce message latency of the intra-grid 2D communication on CPU clusters. Finally, we leverage GPU-initiated one-sided communication to implement the communication trees on GPU clusters. With these nested inter- and intra-grid communication optimization strategies, the proposed 3D SpTRSV algorithm can attain up to 3.45x speedups compared to the baseline 3D SpTRSV algorithm using up to 2048 Cori Haswell CPU cores. In addition, the proposed GPU 3D SpTRSV algorithm can achieve up to 6.5x speedups compared to the proposed CPU 3D SpTRSV algorithm with Pz up to 64. Finally it is remarkable that the proposed GPU 3D SpTRSV can scale to 256 GPUs using the Perlmutter system while the existing 2D SpTRSV algorithm can only scale up to 4 GPUs.

Sao, Piyush↗