Search NASA⌕ Search

SEARCH · Search NASA

Results for “Cloud Architectures”

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 145 records · Page 8
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Virtual Log-Structured Storage for High-Performance Streaming

Over the past decade, given the higher number of data sources (e.g., Cloud applications, Internet of things) and critical business demands, Big Data transitioned from batch-oriented to real-time analytics. Stream storage systems, such as Apache Kafka, are well known for their increasing role in real-time Big Data analytics. For scalable stream data ingestion and processing, they logically split a data stream topic into multiple partitions. Stream storage systems keep multiple data stream copies to protect against data loss while implementing a stream partition as a replicated log. This architectural choice enables simplified development while trading cluster size with performance and the number of streams optimally managed. This paper introduces a shared virtual log-structured storage approach for improving the cluster throughput when multiple producers and consumers write and consume in parallel data streams. Stream partitions are associated with shared replicated virtual logs transparently to the user, effectively separating the implementation of stream partitioning (and data ordering) from data replication (and durability). We implement the virtual log technique in the KerA stream storage system. When comparing with Apache Kafka, KerA improves the cluster ingestion throughput by up to 4x when multiple producers write over hundreds of data streams.

consistent stream ordering↗
DOE OSTI2025

DeepLynx Ecosystem 2025

Poor data integration and governance continue to plague complex engineering projects, resulting in missed cost, schedule, and performance targets. Departments operate in isolated systems with manual data exchange, creating fragmented information that compounds errors and leads to significant delays and cost overruns. The DeepLynx ecosystem addresses these challenges through an open-source, modular data management platform that transforms fragmented project data into an integrated digital thread. Built on a federated microservice architecture, the ecosystem comprises seven specialized tools centered around DeepLynx Nexus, a unified data catalog with hierarchical organization and graph-based navigation capabilities. The ecosystem includes: DeepLynx Stream for real-time timeseries data ingestion from industrial sources; DeepLynx Ingest for governed data uploads with formal review workflows; DeepLynx Lattice for ontology-based entity and relationship extraction; DeepLynx Run for workflow orchestration and secure AI/ML compute; DeepLynx Visualize for 3D digital twin visualization; and DeepLynx Insight for AI-assisted document analysis with traceable, grounded responses. Deployable in cloud, on-premise, or hybrid environments using containerized Docker applications and Helm charts, the DeepLynx ecosystem provides flexible infrastructure that adapts to organizational requirements. By consolidating project data into a unified data lake with role-based access controls and OAuth2 authentication, DeepLynx enables digital thread and digital twin capabilities that improve decision-making, reduce risk, and support complex engineering workflows throughout the project lifecycle.

42 - ENGINEERING↗
DOE OSTI2021

Collaborative Research: Improved Efficiency and Coupling of the Radiation Code in the ACME Earth System Model (Final Report)

The complexity of radiative transfer, its importance to the exchange of energy in the climate system, and its high computational cost establishes importance of an accurate and efficient radiative transfer parameterization for climate simulation. Previous work by the proposing team at AER led to the development of the radiation code, RRTMG, which has been widely accepted by the global modeling community as a fast and accurate advancement over the previous generation of radiation codes. It has been in use in the NCAR CESM for many years, and it has been implemented in the initial version the DOE E3SM model. However, its computational cost remains high relative to other components in part due to its complexity and to its inefficient use of modern optimization strategies, and this project helped address this limitation for the code’s application in E3SM. Under other funding, the Investigators of this project led an effort to develop a high-performance broadband radiation code, called RTE+RRTMGP, which is a completely restructured code that will take advantage of modern computational capabilities to enhance its performance while retaining the strengths and accuracy of the original code. Designed to perform over a range of computer architectures, RTE+RRTMGP makes extensive use of Fortran 2003 features to improve both its efficiency and its use of memory. RTE+RRTMGP is expected to be adopted in the next generation of RTE+RRTMGP. This project allowed for advancements to the code’s computational capabilities and adding new and enhanced features. This included revising RTE+RRTMGP to run on GPU processors, analysis of and improvements to the code’s timing, modifying the gas optics in RRTMGP to increase the code’s accuracy, extensive validation of computed fluxes, heating rates and forcings, generating a cloud optical property and vertical sampling capabilities for RTE+RRTMGP, implementing a fast and accurate longwave scattering capability, and groundwork for the inclusion of a capability to specify solar variability. Many of the accomplishment in this project necessitated significant collaboration with the E3SM development team. The result of this project was optimization of a key physical component (radiative transfer calculations) of E3SM, directly supporting E3SM’s overarching global modeling objectives. More broadly, this project provided overall advancements in the use of radiative transfer calculations in atmospheric modeling and simulation, particularly for climate.

58 GEOSCIENCES↗
DOE OSTI2023

Cloud Atlas: An Interactive Apparatus for Air Quality Awareness

This paper introduces a research project of a kinetic apparatus that translates collected air quality data to users. The kinetic mechanical skeleton is embedded with processors and electronics, and constantly changes its form according to air quality data collected online via a web crawler. By translating invisible environmental conditions, such as air quality data into morphological, colorful, and acoustic changes for the device, the participants can be aware of the phenomenon of air pollution in real time. The interac- tions between participants' movements and the changes of the device imply how human behaviors can affect the state of the atmosphere. As a result, the device evokes public awareness and participation in environmental issues through tangible experiences. This research aims to explore a new kind of responsive architecture that acts as a medium between invisible environments and inhabitants.

Zhao, Xiaoxiao↗
DOE OSTI2021

An Integrated View of Hydraulic Induced Fracture Geometry in Hydraulic Fracture Test Site 2

Until relatively recently, most of our understanding about the geometry of hydraulically induced fractures in unconventional reservoirs has been inferred by patterns observed via microseismic monitoring. Important information, such as fracture-azimuth, fracture-length, height-growth and other geometric data has been determined from microseismic “event clouds”, which in turn has been used to interpret a stimulated reservoir volume (SRV). It is important to note that relatively limited microseismic results have been acquired in conjunction with other diagnostics and validated in a consistent manner. The Hydraulic Fracture Test Site 2 (HFTS2) in the Permian Delaware Basin provides a unique opportunity to compare the frac geometry interpretations derived from multiple frac diagnostic tools and at multiple scales. The integrated view that emerges from all the frac diagnostic tools is that in HFTS2 the geometries of hydraulically induced fractures are not random and with a highly complex branching architecture. In fact, the fracs are mostly vertical, parallel plana domains, and both vertically and horizontally asymmetrical. The hydraulically induced fractures have a very consistent azimuth with a strike that matches SHmax. At the distance corresponding to the spacing between nearby wells in HFTS2, ~ 660 ft, the pattern of strain interceptions observed from Low Frequency Distributed Acoustic Sensing (LF-DAS) shows far-field dimensions that are consistent with the stage length dimensions at the stimulated well. In general, larger stages with higher number of clusters create larger number of interceptions and wider stimulated intervals than that of shorter stages with the same cluster to cluster spacing and fewer clusters. The hydraulically induced fractures also show a clear tendency to preferentially grow upwards as shown by the microseismic data, by the relative lower number of LF-DAS interceptions in nearby deeper wells and finally by the observed LF-DAS strain patterns in the vertical observation well. In the near field, the geometries of the fracture domains are also consistent. As shown by the distributed strain measurements obtained during production via the monitoring of Rayleigh Frequency Shift (RFS), each cluster has a separate, non-overlapping, frac-zone-domain that are generally centered around the locations of each perforation cluster. Overall, the hydraulically induced fractures in HFTS2 are interpreted as occurring in swarms associated with individual clusters, each with its own unique geometry but with similar predictable geometric and propagation tendencies. The comprehensive frac diagnostic program in HFTS2, the well-pad layout that permits the investigation of the Downloaded from http://onepetro.org/URTECONF/proceedings-pdf/21URTC/1-21URTC/D011S018R003/2477511/urtec-2021-5396-ms.pdf/1 by Carol Worster on 28 February 2022 URTeC 5396 2 impact that nearby producing wells has on frac geometry, together with an innovative Design of Experiment (DofE) that includes sequential-fracing, consistent cluster-to-cluster distances and few single-perforation cluster stages, has allowed us to gain a unique perspective about the geometry of the induced hydraulic fractures on this pad. Finally, the HFTS2 dataset has also provided us with a better understanding of the applicability and limitations of the different diagnostic tools.

58 GEOSCIENCES↗
DOE OSTI2023

Experimental Observations of the Topology of Convolutional Neural Network Activations

Topological data analysis (TDA) is a branch of computational mathematics, bridging algebraic topology and data science, that provides compact, noise-robust representations of complex structures. Deep neural networks (DNNs) learn millions of parameters associated with a series of transformations defined by the model architecture resulting in high-dimensional, difficult to interpret internal representations of input data. As DNNs become more ubiquitous across multiple sectors of our society, there is increasing recognition that mathematical methods are needed to aid analysts, researchers, and practitioners in understanding and interpreting how these models' internal representations relate to the final classification. In this paper we apply cutting edge techniques from TDA with the goal of gaining insight towards interpretability of convolutional neural networks used for image classification. We use two common TDA approaches to explore several methods for modeling hidden layer activations as high-dimensional point clouds, and provide experimental evidence that these point clouds capture valuable structural information about the model's process. First, we demonstrate that a distance metric based on persistent homology can be used to quantify meaningful differences between layers and discuss these distances in the broader context of existing representational similarity metrics for neural network interpretability. Second, we show that a mapper graph can provide semantic insight as to how these models organize hierarchical class knowledge at each layer. These observations demonstrate that TDA is a useful tool to help deep learning practitioners unlock the hidden structures of their models.

topological data analysis, deep learning↗
DOE OSTI2025

Zero Trust Strategies for Chemical, Biological, Radiological, and Nuclear Detection Systems: D.1 Cyber Scenarios

The evolving landscape of cybersecurity necessitates a paradigm shift to a Zero Trust (ZT) model, which assumes breaches and continuously verifies trust. This approach reshapes how trust boundaries are established, focusing on identities, devices, networks, applications, and data, rather than solely relying on perimeter defenses such as firewalls. Central to this transformation is the National Institute of Standards and Technology's (NIST) Special Publication 800-207, outlining the Zero Trust Architecture (ZTA), along with Executive Order 14028, which mandates federal agencies to adopt ZT principles. Complementary to these efforts, the Cybersecurity and Infrastructure Security Agency (CISA) developed the Zero Trust Maturity Model (ZTMM), providing a framework with five pillars and three cross-cutting capabilities to guide agencies toward enhanced cybersecurity maturity. In support of these initiatives, the DHS Countering Weapons of Mass Destruction Office (CWMD) is applying ZT principles to secure Chemical, Biological, Radiological, and Nuclear (CBRN) detection systems. Recognizing the diverse deployment models and network connectivity of these systems—from stationary, non-networked units to mobile, cloud-connected devices—the Pacific Northwest National Laboratory (PNNL) is developing cybersecurity scenarios specifically for CBRN environments. These scenarios examine various configurations and technological capabilities, offering insights into the application of ZTMM pillars in enhancing the security postures of CBRN devices. The cybersecurity scenarios presented by PNNL are hypothetical, crafted to explore theoretical situations and stimulate discussion on the potential use or compromise of CBRN detection systems in varied contexts. These narratives are illustrative and do not reference any real events or actual networks. Instead, they employ generalized reference models to highlight concepts and potential issues within CBRN security, focusing on how Zero Trust strategies can be adapted to address these challenges effectively.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗
DOE OSTI2021

Rahasak—Scalable blockchain architecture for enterprise applications

Blockchain-based decentralized infrastructure has been adapted in various industries to handle the sensitive data in a privacy-preserving manner without trusting third parties. However, integrating state-of-the-art blockchain platforms with the scalable, enterprise-level applications result in several challenges. Current blockchain platforms do not support high transaction throughput, lack high scalability, and cannot provide real-time transaction processing and back-pressure operation handling in high transaction throughput applications(e.g Big data, IoT). In this paper, we propose a novel permissioned blockchain platform “Rahasak” for highly scalable, enterprise applications. Rahasak blockchain adopts the Apache Kafka-based consensus on top of a “Validate-Execute-Group” blockchain architecture to handle realtime transaction execution on the blockchain. The architecture is equipped with a functional programming and actor-based smart contract platform that enables concurrent execution of transactions in the blockchain. Rahasak supports high transaction throughput, high scalability, concurrent transaction execution, data analytics features. Finally, with Rahasak, we make blockchain more scalable, secure, structured and meaningful for further data analytics.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

iDDS: intelligent distributed dispatch and scheduling for workflow orchestration

The intelligent distributed dispatch and scheduling (iDDS) service is a versatile workflow orchestration system designed for large-scale, distributed scientific computing. iDDS extends traditional workload and data management by integrating data-aware execution, conditional logic, and programmable workflows, enabling automation of complex and dynamic processing pipelines. Originally developed for the ATLAS experiment at the large hadron collider, iDDS has evolved into an experiment-agnostic platform that supports both template-driven workflows and a Function-as-a-Task model for Python-based orchestration. This paper presents the architecture and core components of iDDS, highlighting its scalability, modular message-driven design, and integration with systems such as PanDA and Rucio. We demonstrate its versatility through real-world use cases: fine-grained tape resource optimization for ATLAS, orchestration of large Directed Acyclic Graph (DAG) workflows for the Rubin Observatory, distributed hyperparameter optimization for machine learning applications, active learning for physics analyses, and AI-assisted detector design at the electron–ion collider. By unifying workload scheduling, data movement, and adaptive decision-making, iDDS reduces operational overhead and enables reproducible, high-throughput workflows across heterogeneous infrastructures. We conclude with current challenges and future directions, including interactive, cloud-native, and serverless workflow support.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Generalizable Web User Interface for Scalable and Streamlined Deployment of Building Energy Management Systems in Small and Medium-Sized Commercial Buildings

Small and medium-sized commercial buildings (SMCBs) comprise 94% of US commercial buildings yet face significant barriers to implementing building energy management systems despite advances in smart device technology. Existing solutions present critical limitations: cloud-based API solutions simplify deployment but create vendor lock-in constraints; commercial integrated software solutions ensure compatibility via standardized protocols but require substantial cost and technical expertise; open-source IoT platforms offer cost-effective vendor independence but provide insufficient standardized protocol support for commercial building automation. This research presents a generalizable web user interface framework that bridges the gap between evolving smart device capabilities and lagging software infrastructure for SMCBs. The proposed system integrates VOLTTRON open-source middleware with an automated configuration converter that transforms unified specifications written in YAML, a human-readable data-serialization format, into system-specific files, streamlining manual setup processes. The vendor-agnostic architecture supports industry-standard protocols (BACnet and Modbus) and semantic building models while providing adaptive web interfaces that dynamically adjust to various building configurations. Demonstrations through simulation-based testing and a field deployment show automatic interface adaptation across heterogeneous HVAC systems and multizone monitoring. The automated configuration converter also substantially reduces labor-intensive setup.

Chung, Jihoon [ORNL] (ORCID:0000000184880815)↗
DOE OSTI2024

A Comparison between Invariant and Equivariant Classical and Quantum Graph Neural Networks

Machine learning algorithms are heavily relied on to understand the vast amounts of data from high-energy particle collisions at the CERN Large Hadron Collider (LHC). The data from such collision events can naturally be represented with graph structures. Therefore, deep geometric methods, such as graph neural networks (GNNs), have been leveraged for various data analysis tasks in high-energy physics. One typical task is jet tagging, where jets are viewed as point clouds with distinct features and edge connections between their constituent particles. The increasing size and complexity of the LHC particle datasets, as well as the computational models used for their analysis, have greatly motivated the development of alternative fast and efficient computational paradigms such as quantum computation. In addition, to enhance the validity and robustness of deep networks, we can leverage the fundamental symmetries present in the data through the use of invariant inputs and equivariant layers. In this paper, we provide a fair and comprehensive comparison of classical graph neural networks (GNNs) and equivariant graph neural networks (EGNNs) and their quantum counterparts: quantum graph neural networks (QGNNs) and equivariant quantum graph neural networks (EQGNN). The four architectures were benchmarked on a binary classification task to classify the parton-level particle initiating the jet. Based on their area under the curve (AUC) scores, the quantum networks were found to outperform the classical networks. However, seeing the computational advantage of quantum networks in practice may have to wait for the further development of quantum technology and its associated application programming interfaces (APIs).

Forestano, Roy T. (ORCID:0000000203552076)↗
DOE OSTI2020

Joint Genome Institute Analysis Workflow Service (JAWS) v2.7

The U.S. Department of Energy Joint Genome Institute (JGI) has developed the JGI Analysis Workflow Service (JAWS) as a distributed framework to run computational workflows across diverse high-performance computing (HPC) and cloud environments. JAWS enhances the reusability, scalability, and robustness of scientific workflows by orchestrating data movement, code execution, and results retrieval across multiple DOE facilities. At its core, JAWS integrates the Cromwell workflow engine to run workflows expressed in the Workflow Description Language (WDL), ensuring portability and interoperability. To provide consistent runtime environments, JAWS employs container technologies such as Shifter, Apptainer, and Docker. Workflow tasks are managed via HTCondor on HPC backends, while Globus ensures secure and efficient data transfer between sites. JAWS is deployed as a multi-site workflow manager across national laboratory computing facilities, with dedicated instances supporting community projects such as the National Microbiome Data Collaborative (NMDC) and KBase. This distributed, service-oriented architecture enables users to "write once, run anywhere," providing scalable, production-quality workflow execution.

Kirton, Edward↗
DOE OSTI2023

The globus compute dataset: An open function-as-a-service dataset from the edge to the cloud

Here we present a unique function-as-a-service (FaaS) dataset capturing the use of the Globus Compute (previously funcX) platform. Globus Compute implements a federated model via which users may deploy endpoints on arbitrary remote computers, from the edge to high performance computing (HPC) cluster, and they may then invoke Python functions on those endpoints via a reliable cloud -hosted service. The dataset covers 31 weeks and includes 2121472 task submissions from 252 users executed on 580 remote computing endpoints. It includes 277386 registered functions. We describe the dataset and various observations, some that are similar to other FaaS datasets, for example, that 74% of tasks run for less than 1 s, and some that are unique to Globus Compute, for example, that endpoints are used in different ways and that the majority of functions are related to scientific computing and machine learning. To the best of our knowledge, this dataset represents the first federated FaaS dataset that includes user workloads, distributed computing endpoints, and analysis of registered function bodies. We expect the dataset to be useful for researching FaaS architectures, workload modeling, container warming, and other distributed computing architectures.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Quantum computation of silicon electronic band structure

Development of quantum architectures during the last decade has inspired hybrid classical–quantum algorithms in physics and quantum chemistry that promise simulations of fermionic systems beyond the capability of modern classical computers, even before the era of quantum computing fully arrives. Strong research efforts have been recently made to obtain minimal depth quantum circuits which could accurately represent chemical systems. Here, we show that unprecedented methods used in quantum chemistry, designed to simulate molecules on quantum processors, can be extended to calculate properties of periodic solids. In particular, we present minimal depth circuits implementing the variational quantum eigensolver algorithm and successfully use it to compute the band structure of silicon on a quantum machine for the first time. We are convinced that the presented quantum experiments performed on cloud-based platforms will stimulate more intense studies towards scalable electronic structure computation of advanced quantum materials.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Coding the Computing Continuum: Fluid Function Execution in Heterogeneous Computing Environments

Advances in network technologies have greatly decreased barriers to accessing physically distributed computers. This newfound accessibility coincides with increasing hardware specialization, creating exciting new opportunities to dispatch workloads to the best resource for a specific purpose, rather than those that are closest or most easily accessible. We present Delta, a service designed to intelligently schedule function-based workloads across a distributed set of heterogeneous computing resources. Delta implements an extensible architecture in which different predictors and scheduling algorithms can be integrated to provide dynamically evolving estimates of function execution times on different resources-estimates that can be used to determine the most appropriate location for execution. We describe predictors for function runtime, data transfer time, and cold-start resource provisioning and configuration delay; dynamic learning methods that update predictor models over time; and scheduling strategies that take into account both function and endpoint information. We show that these methods can halve workload makespan when compared with a strategy that selects the fastest resource, and decrease makespan by a factor of five when compared to a round robin strategy, when deployed on a heterogeneous testbed with resources ranging from a Raspberry Pi to a GPU node in an academic cloud.

Computing continuum↗
DOE OSTI2024

Anticipating Optical Availability in Hybrid RF/FSO Links Using RF Beacons and Deep Learning

Radiofrequency (RF) communications offer reliable but low data rates and energy-inefficient satellite links, while free-space optical (FSO) promises high bandwidth but struggles with disturbances imposed by atmospheric effects. A hybrid RF/FSO architecture aims to achieve optimal reliability along with high data rates for space communications. Accurate prediction of dynamic ground-to-satellite FSO link availability is critical for routing decisions in low-earth orbit constellations. In this paper, we propose a system leveraging ubiquitous RF links to proactively forecast FSO link degradation prior to signal drops below threshold levels. This enables pre-calculation of rerouting to maximally maintain high data rate FSO links throughout the duration of weather effects. We implement a supervised learning model to anticipate FSO attenuation based on the analysis of RF patterns. Through the simulation of a dense lower earth orbit (LEO) satellite constellation, we demonstrate the efficacy of our approach in a simulated satellite network, highlighting the balance between predictive accuracy and prediction duration. An emulated cloud attenuation model is proposed to provide insight into the temporal profiles of RF signals and their correlation to FSO channel dynamics. Our investigation sheds light on the trade-offs between prediction horizon and accuracy arising from RF beacon numbers and proximity.

FSO availability↗
DOE OSTI2020

Deep Generative Models that Solve PDEs: Distributed Computing for Training Large Data-Free Models

Recent progress in scientific machine learning (SciML) has opened up the possibility of training novel neural network architectures that solve complex partial differential equations (PDEs). Several (nearly data free) approaches have been recently reported that successfully solve PDEs, with examples including deep feed forward networks, generative networks, and deep encoder-decoder networks. However, practical adoption of these approaches is limited by the difficulty in training these models, especially to make predictions at large output resolutions (≥1024×1024). Here we report on a software framework for data parallel distributed deep learning that resolves the twin challenges of training these large SciML models - training in reasonable time as well as distributing the storage requirements. Our framework provides several out of the box functionality including (a) loss integrity independent of number of processes, (b) synchronized batch normalization, and (c) distributed higher-order optimization methods. We show excellent scalability of this framework on both cloud as well as HPC clusters, and report on the interplay between bandwidth, network topology and bare metal vs cloud. We deploy this approach to train generative models of sizes hitherto not possible, showing that neural PDE solvers can be viably trained for practical applications. We also demonstrate that distributed higher-order optimization methods are 2-3× faster than stochastic gradient-based methods and provide minimal convergence drift with higher batch-size.

PDEs↗
DOE OSTI2023

Transforming microseismic clouds into near real-time visualization of the growing hydraulic fracture

SUMMARY Microseismic observations during unconventional reservoir stimulation are typically seen as a proxy for clusters of hydraulic fractures and the extent of the stimulated reservoir. Such straightforward interpretation is often misleading and fails to provide a physically reasonable image of the fracturing process. This paper demonstrates the application of a physics-based machine learning algorithm which enables a rapid and accurate fracture mapping from the microseismic data. Our training and validation data set relies on a history-matched geomechanical modelling workflow implemented in GEOS software for the Hydraulic Fracturing Test Site 1 (HFTS-1) project. For this study we augmented the simulated fracture growth through geostatistical modelling of induced seismicity, so that the synthetic microseismic catalogue matches the main statistical properties of the field observations. We formulated the problem of mapping the actual fracture in the clutter of events to parallel common video segmentation workflows: several past video frames (microseismic density snapshots) are passed through a deep convolutional network to classify whether a given voxel is associated with a fracture or intact rock. We found that for accurate fracture mapping, the network’s input and architecture must be augmented to incorporate the fluid injection parameters (pressure, rate, concentration of proppant, and location of the perforation within the cluster). The error rate for the network reached as little as 10 per cent of the fracture area, while a conventional microseismic interpretation approach yielded ∼300 per cent. Our approach also yields must faster predictions than conventional methods (minutes instead of weeks), and could enable engineers to make rapid decisions regarding engineering parameters (pumping rate, viscosity) in real time during stimulation.

58 GEOSCIENCES↗