In software engineering, class diagrams are often used to describe the system＇s class structures in Unified Modelling Language （UML）. A class diagram, as a graph, is a collection of static declarative model element...In software engineering, class diagrams are often used to describe the system＇s class structures in Unified Modelling Language （UML）. A class diagram, as a graph, is a collection of static declarative model elements, such as classes, interfaces, and the relationships of their connections with each other. In this paper, class graphs axe examined within several Java software systems provided by Sun and IBM, and some new features are found. For a large-scale Java software system, its in-degree distribution tends to an exponential distribution, while its out-degree and degree distributions reveal the power-law behaviour. And then a directed preferential-random model is established to describe the corresponding degree distribution features and evolve large-scale Java software systems.展开更多
To search for the Design Patterns’ influence on the software, the paper abstracts the feature models of 9 kinds of classic exiting design patterns among the 23 kinds and describes the features with algorithm language...To search for the Design Patterns’ influence on the software, the paper abstracts the feature models of 9 kinds of classic exiting design patterns among the 23 kinds and describes the features with algorithm language. Meanwhile, searching for the specific structure features in the network, the paper designs 9 matching algorithms of the 9 kinds design patterns mentioned above to research on the structure of the design patterns in the software network. At last, the paper analyzes the evolving trends of the software scale and the application frequency of the 9 kinds of design patterns as the software evolves, and search for the rules how these design patterns are applied into 4 kinds of typical software.展开更多
Large-scale software systems,which are the most sophisticated human-designed objects,play more and more important role in our daily life.Consequently effective analysis for large-scale software has become an urgent pr...Large-scale software systems,which are the most sophisticated human-designed objects,play more and more important role in our daily life.Consequently effective analysis for large-scale software has become an urgent problem to be solved with the increasing issues of software security and the continuous expansion of software applications scope.For the characteristics of large scale and complex structure in large-scale software,the traditional software analysis techniques are difficult to be used.With the problem of difficulty in presentation,storage and low efficiency in the process of large-scale software analysis,the visualization analysis framework for large-scale software based on software network,named SoNet,is proposed with the combination of complex network theory and program slicing technique.Constraint logic attributes of the programs will be obtained through source code parsing.Then we will construct a global view by the theory of complex network after extracting software structure and behavior,improving user’s perception of software architecture in a macro perspective.Use case slicing will be realized combined with Redis cluster,and accessibility analysis when given a keyword to be analyzed.We evaluate our prototype implementation on an open source software project named SoundSea in Github,and the results suggest that our approach can realize the analysis for large-scale software.展开更多
The advent of parametric design has resulted in a marked increase in the complexity of building.Unfortunately,traditional construction methods make it difficult to meet the needs.Therefore,construction robots have bec...The advent of parametric design has resulted in a marked increase in the complexity of building.Unfortunately,traditional construction methods make it difficult to meet the needs.Therefore,construction robots have become a pivotal production tool in this context.Since the arm span of a single robot usually does not exceed 3 meters,it is not competent for producing large-scale building components.Accordingly,the extension of the robot,s working range is often achieved by external axes.Nevertheless,the coupling control of external axes and robots and their kinematic solution have become key challenges.The primary technical difficulties include customized construction robots,automatic solutions for external axes,fixed axis joints,and specific motion mode control.This paper proposes solutions to these difficulties,introduces the relevant basic concepts and algorithms in detail,and encapsulates these robotics principles and algorithm processes into the Grasshopper plug-in commonly used by architects to form the FURobot software platform.This platform effectively solves the above problems,lowers the threshold for architects,and improves production efficiency.The effectiveness of the algorithm and software in this paper is verified through simulation experiments.展开更多
1.Introduction Climate change mitigation pathways aimed at limiting global anthropogenic carbon dioxide(CO_(2))emissions while striving to constrain the global temperature increase to below 2℃—as outlined by the Int...1.Introduction Climate change mitigation pathways aimed at limiting global anthropogenic carbon dioxide(CO_(2))emissions while striving to constrain the global temperature increase to below 2℃—as outlined by the Intergovernmental Panel on Climate Change(IPCC)—consistently predict the widespread implementation of CO_(2)geological storage on a global scale.展开更多
Software security poses substantial risks to our society because software has become part of our life. Numerous techniques have been proposed to resolve or mitigate the impact of software security issues. Among them, ...Software security poses substantial risks to our society because software has become part of our life. Numerous techniques have been proposed to resolve or mitigate the impact of software security issues. Among them, software testing and analysis are two of the critical methods, which significantly benefit from the advancements in deep learning technologies. Due to the successful use of deep learning in software security, recently,researchers have explored the potential of using large language models(LLMs) in this area. In this paper, we systematically review the results focusing on LLMs in software security. We analyze the topics of fuzzing, unit test, program repair, bug reproduction, data-driven bug detection, and bug triage. We deconstruct these techniques into several stages and analyze how LLMs can be used in the stages. We also discuss the future directions of using LLMs in software security, including the future directions for the existing use of LLMs and extensions from conventional deep learning research.展开更多
Spectrum-based fault localization (SBFL) generates a ranked list of suspicious elements by using the program execution spectrum, but the excessive number of elements ranked in parallel results in low localization accu...Spectrum-based fault localization (SBFL) generates a ranked list of suspicious elements by using the program execution spectrum, but the excessive number of elements ranked in parallel results in low localization accuracy. Most researchers consider intra-class dependencies to improve localization accuracy. However, some studies show that inter-class method call type faults account for more than 20%, which means such methods still have certain limitations. To solve the above problems, this paper proposes a two-phase software fault localization based on relational graph convolutional neural networks (Two-RGCNFL). Firstly, in Phase 1, the method call dependence graph (MCDG) of the program is constructed, the intra-class and inter-class dependencies in MCDG are extracted by using the relational graph convolutional neural network, and the classifier is used to identify the faulty methods. Then, the GraphSMOTE algorithm is improved to alleviate the impact of class imbalance on classification accuracy. Aiming at the problem of parallel ranking of element suspicious values in traditional SBFL technology, in Phase 2, Doc2Vec is used to learn static features, while spectrum information serves as dynamic features. A RankNet model based on siamese multi-layer perceptron is constructed to score and rank statements in the faulty method. This work conducts experiments on 5 real projects of Defects4J benchmark. Experimental results show that, compared with the traditional SBFL technique and two baseline methods, our approach improves the Top-1 accuracy by 262.86%, 29.59% and 53.01%, respectively, which verifies the effectiveness of Two-RGCNFL. Furthermore, this work verifies the importance of inter-class dependencies through ablation experiments.展开更多
Software-defined networking(SDN)is an innovative paradigm that separates the control and data planes,introducing centralized network control.SDN is increasingly being adopted by Carrier Grade networks,offering enhance...Software-defined networking(SDN)is an innovative paradigm that separates the control and data planes,introducing centralized network control.SDN is increasingly being adopted by Carrier Grade networks,offering enhanced networkmanagement capabilities than those of traditional networks.However,because SDN is designed to ensure high-level service availability,it faces additional challenges.One of themost critical challenges is ensuring efficient detection and recovery from link failures in the data plane.Such failures can significantly impact network performance and lead to service outages,making resiliency a key concern for the effective adoption of SDN.Since the recovery process is intrinsically dependent on timely failure detection,this research surveys and analyzes the current literature on both failure detection and recovery approaches in SDN.The survey provides a critical comparison of existing failure detection techniques,highlighting their advantages and disadvantages.Additionally,it examines the current failure recovery methods,categorized as either restoration-based or protection-based,and offers a comprehensive comparison of their strengths and limitations.Lastly,future research challenges and directions are discussed to address the shortcomings of existing failure recovery methods.展开更多
Software-related security aspects are a growing and legitimate concern,especially with 5G data available just at our palms.To conduct research in this field,periodic comparative analysis is needed with the new techniq...Software-related security aspects are a growing and legitimate concern,especially with 5G data available just at our palms.To conduct research in this field,periodic comparative analysis is needed with the new techniques coming up rapidly.The purpose of this study is to review the recent developments in the field of security integration in the software development lifecycle(SDLC)by analyzing the articles published in the last two decades and to propose a way forward.This review follows Kitchenham’s review protocol.The review has been divided into three main stages including planning,execution,and analysis.From the selected 100 articles,it becomes evident that need of a collaborative approach is necessary for addressing critical software security risks(CSSRs)through effective risk management/estimation techniques.Quantifying risks using a numeric scale enables a comprehensive understanding of their severity,facilitating focused resource allocation and mitigation efforts.Through a comprehensive understanding of potential vulnerabilities and proactive mitigation efforts facilitated by protection poker,organizations can prioritize resources effectively to ensure the successful outcome of projects and initiatives in today’s dynamic threat landscape.The review reveals that threat analysis and security testing are needed to develop automated tools for the future.Accurate estimation of effort required to prioritize potential security risks is a big challenge in software security.The accuracy of effort estimation can be further improved by exploring new techniques,particularly those involving deep learning.It is also imperative to validate these effort estimation methods to ensure all potential security threats are addressed.Another challenge is selecting the right model for each specific security threat.To achieve a comprehensive evaluation,researchers should use well-known benchmark checklists.展开更多
Link failure is a critical issue in large networks and must be effectively addressed.In software-defined networks(SDN),link failure recovery schemes can be categorized into proactive and reactive approaches.Reactive s...Link failure is a critical issue in large networks and must be effectively addressed.In software-defined networks(SDN),link failure recovery schemes can be categorized into proactive and reactive approaches.Reactive schemes have longer recovery times while proactive schemes provide faster recovery but overwhelm the memory of switches by flow entries.As SDN adoption grows,ensuring efficient recovery from link failures in the data plane becomes crucial.In particular,data center networks(DCNs)demand rapid recovery times and efficient resource utilization to meet carrier-grade requirements.This paper proposes an efficient Decentralized Failure Recovery(DFR)model for SDNs,meeting recovery time requirements and optimizing switch memory resource consumption.The DFR model enables switches to autonomously reroute traffic upon link failures without involving the controller,achieving fast recovery times while minimizing memory usage.DFR employs the Fast Failover Group in the OpenFlow standard for local recovery without requiring controller communication and utilizes the k-shortest path algorithm to proactively install backup paths,allowing immediate local recovery without controller intervention and enhancing overall network stability and scalability.DFR employs flow entry aggregation techniques to reduce switch memory usage.Instead of matching flow entries to the destination host’s MAC address,DFR matches packets to the destination switch’s MAC address.This reduces the switches’Ternary Content-Addressable Memory(TCAM)consumption.Additionally,DFR modifies Address Resolution Protocol(ARP)replies to provide source hosts with the destination switch’s MAC address,facilitating flow entry aggregation without affecting normal network operations.The performance of DFR is evaluated through the network emulator Mininet 2.3.1 and Ryu 3.1 as SDN controller.For different number of active flows,number of hosts per edge switch,and different network sizes,the proposed model outperformed various failure recovery models:restoration-based,protection by flow entries,protection by group entries and protection by Vlan-tagging model in terms of recovery time,switch memory consumption and controller overhead which represented the number of flow entry updates to recover from the failure.Experimental results demonstrate that DFR achieves recovery times under 20 milliseconds,satisfying carrier-grade requirements for rapid failure recovery.Additionally,DFR reduces switch memory usage by up to 95%compared to traditional protection methods and minimizes controller load by eliminating the need for controller intervention during failure recovery.Theresults underscore the efficiency and scalability of the DFR model,making it a practical solution for enhancing network resilience in SDN environments.展开更多
The titanium alloy strut serves as a key load-bearing component of aircraft landing gear,typically manufactured via forging.The friction condition has important influence on material flow and cavity filling during the...The titanium alloy strut serves as a key load-bearing component of aircraft landing gear,typically manufactured via forging.The friction condition has important influence on material flow and cavity filling during the forging process.Using the previously optimized shape and initial position of preform,the influence of the friction condition(friction factor m=0.1–0.3)on material flow and cavity filling was studied by numerical method with a shear friction model.A novel filling index was defined to reflect material flow into left and right flashes and zoom in on friction-induced results.The results indicate that the workpiece moves rigidly to the right direction,with the displacement decreasing as m increases.When m<0.18,the underfilling defect will occur in the left side of strut forging,while overflow occurs in the right forging die cavity.By combining the filling index and analyses of material flow and filling status,a reasonable friction factor interval of m=0.21–0.24 can be determined.Within this interval,the cavity filling behavior demonstrates robustness,with friction fluctuations exerting minimal influence.展开更多
In this work,we present a parallel implementation of radiation hydrodynamics coupled with particle transport,utilizing software infrastructure JASMIN(J Adaptive Structured Meshes applications INfrastructure)which enca...In this work,we present a parallel implementation of radiation hydrodynamics coupled with particle transport,utilizing software infrastructure JASMIN(J Adaptive Structured Meshes applications INfrastructure)which encapsulates high-performance technology for the numerical simulation of complex applications.Two serial codes,radiation hydrodynamics RH2D and particle transport Sn2D,have been integrated into RHSn2D on JASMIN infrastructure,which can efficiently use thousands of processors to simulate the complex multi-physics phenomena.Moreover,the non-conforming processors strategy has ensured RHSn2D against the serious load imbalance between radiation hydrodynamics and particle transport for large scale parallel simulations.Numerical results show that RHSn2D achieves a parallel efficiency of 17.1%using 90720 cells on 8192 processors compared with 256 processors in the same problem.展开更多
Quantum software development utilizes quantum phenomena such as superposition and entanglement to address problems that are challenging for classical systems.However,it must also adhere to critical quantum constraints...Quantum software development utilizes quantum phenomena such as superposition and entanglement to address problems that are challenging for classical systems.However,it must also adhere to critical quantum constraints,notably the no-cloning theorem,which prohibits the exact duplication of unknown quantum states and has profound implications for cryptography,secure communication,and error correction.While existing quantum circuit representations implicitly honor such constraints,they lack formal mechanisms for early-stage verification in software design.Addressing this constraint at the design phase is essential to ensure the correctness and reliability of quantum software.This paper presents a formal metamodeling framework using UML-style notation and and Object Constraint Language(OCL)to systematically capture and enforce the no-cloning theorem within quantum software models.The proposed metamodel formalizes key quantum concepts—such as entanglement and teleportation—and encodes enforceable invariants that reflect core quantum mechanical laws.The framework’s effectiveness is validated by analyzing two critical edge cases—conditional copying with CNOT gates and quantum teleportation—through instance model evaluations.These cases demonstrate that the metamodel can capture nuanced scenarios that are often mistaken as violations of the no-cloning theorem but are proven compliant under formal analysis.Thus,these serve as constructive validations that demonstrate the metamodel’s expressiveness and correctness in representing operations that may appear to challenge the no-cloning theorem but,upon rigorous analysis,are shown to comply with it.The approach supports early detection of conceptual design errors,promoting correctness prior to implementation.The framework’s extensibility is also demonstrated by modeling projective measurement,further reinforcing its applicability to broader quantum software engineering tasks.By integrating the rigor of metamodeling with fundamental quantum mechanical principles,this work provides a structured,model-driven approach that enables traditional software engineers to address quantum computing challenges.It offers practical insights into embedding quantum correctness at the modeling level and advances the development of reliable,error-resilient quantum software systems.展开更多
In recent years,with the rapid development of software systems,the continuous expansion of software scale and the increasing complexity of systems have led to the emergence of a growing number of software metrics.Defe...In recent years,with the rapid development of software systems,the continuous expansion of software scale and the increasing complexity of systems have led to the emergence of a growing number of software metrics.Defect prediction methods based on software metric elements highly rely on software metric data.However,redundant software metric data is not conducive to efficient defect prediction,posing severe challenges to current software defect prediction tasks.To address these issues,this paper focuses on the rational clustering of software metric data.Firstly,multiple software projects are evaluated to determine the preset number of clusters for software metrics,and various clustering methods are employed to cluster the metric elements.Subsequently,a co-occurrence matrix is designed to comprehensively quantify the number of times that metrics appear in the same category.Based on the comprehensive results,the software metric data are divided into two semantic views containing different metrics,thereby analyzing the semantic information behind the software metrics.On this basis,this paper also conducts an in-depth analysis of the impact of different semantic view of metrics on defect prediction results,as well as the performance of various classification models under these semantic views.Experiments show that the joint use of the two semantic views can significantly improve the performance of models in software defect prediction,providing a new understanding and approach at the semantic view level for defect prediction research based on software metrics.展开更多
The advent of large language models(LLMs)has made knowledge acquisition and content creation increasingly easier and cheaper,which in turn redefines learning and urges transformation in software engineering education....The advent of large language models(LLMs)has made knowledge acquisition and content creation increasingly easier and cheaper,which in turn redefines learning and urges transformation in software engineering education.To do so,there is a need to understand the impact of LLMs on software engineering education.In this paper,we conducted a preliminary case study on three software requirements engineering classes where students are allowed to use LLMs to assist in their projects.Based on the students’experience,performance,and feedback from a survey conducted at the end of the courses,we characterized the challenges and benefits of applying LLMs in software engineering education.This research contributes to the ongoing discourse on the integration of LLMs in education,emphasizing both their prominent potential and the need for balanced,mindful usage.展开更多
This paper presents our endeavors in developing the large-scale, ultra-high-resolution E3SM Land Model (uELM), specifically designed for exascale computers furnished with accelerators such as Nvidia GPUs. The uELM is ...This paper presents our endeavors in developing the large-scale, ultra-high-resolution E3SM Land Model (uELM), specifically designed for exascale computers furnished with accelerators such as Nvidia GPUs. The uELM is a sophisticated code that substantially relies on High-Performance Computing (HPC) environments, necessitating particular machine and software configurations. To facilitate community-based uELM developments employing GPUs, we have created a portable, standalone software environment preconfigured with uELM input datasets, simulation cases, and source code. This environment, utilizing Docker, encompasses all essential code, libraries, and system software for uELM development on GPUs. It also features a functional unit test framework and an offline model testbed for comprehensive numerical experiments. From a technical perspective, the paper discusses GPU-ready container generations, uELM code management, and input data distribution across computational platforms. Lastly, the paper demonstrates the use of environment for functional unit testing, end-to-end simulation on CPUs and GPUs, and collaborative code development.展开更多
Formalizing complex processes and phenomena of a real-world problem may require a large number of variables and constraints,resulting in what is termed a large-scale optimization problem.Nowadays,such large-scale opti...Formalizing complex processes and phenomena of a real-world problem may require a large number of variables and constraints,resulting in what is termed a large-scale optimization problem.Nowadays,such large-scale optimization problems are solved using computing machines,leading to an enormous computational time being required,which may delay deriving timely solutions.Decomposition methods,which partition a large-scale optimization problem into lower-dimensional subproblems,represent a key approach to addressing time-efficiency issues.There has been significant progress in both applied mathematics and emerging artificial intelligence approaches on this front.This work aims at providing an overview of the decomposition methods from both the mathematics and computer science points of view.We also remark on the state-of-the-art developments and recent applications of the decomposition methods,and discuss the future research and development perspectives.展开更多
This article focuses on the management of large-scale machinery and equipment in highway construction,with the research objective of identifying issues at the management level and exploring more effective management m...This article focuses on the management of large-scale machinery and equipment in highway construction,with the research objective of identifying issues at the management level and exploring more effective management measures.Through practical observation and logical analysis,this article elaborates on the management connotations of large-scale machinery and equipment in highway construction,affirming its management value from different perspectives.On this basis,it carefully analyzes the problems existing in the management of large-scale machinery and equipment,providing a detailed interpretation of issues such as the weak foundation of the equipment management system and the disconnection between equipment selection and configuration from reality.Combining the manifestations of related problems,this article proposes strategies such as strengthening the institutional foundation of equipment management,selecting and configuring equipment based on actual conditions,aiming to provide references for large-scale machinery and equipment management to relevant enterprises.展开更多
The recent upsurge in metro construction emphasizes the necessity of understanding the mechanical performance of metro shield tunnel subjected to the influence of ground fissures.In this study,a largescale experiment,...The recent upsurge in metro construction emphasizes the necessity of understanding the mechanical performance of metro shield tunnel subjected to the influence of ground fissures.In this study,a largescale experiment,in combination with numerical simulation,was conducted to investigate the influence of ground fissures on a metro shield tunnel.The results indicate that the lining contact pressure at the vault increases in the hanging wall while decreases in the footwall,resulting in a two-dimensional stress state of vertical shear and axial tension-compression,and simultaneous vertical dislocation and axial tilt for the segments around the ground fissure.In addition,the damage to curved bolts includes tensile yield,flexural yield,and shear twist,leading to obvious concrete lining damage,particularly at the vault,arch bottom,and hance,indicating that the joints in these positions are weak areas.The shield tunnel orthogonal to the ground fissure ultimately experiences shear failure,suggesting that the maximum actual dislocation of ground fissure that the structure can withstand is approximately 20 cm,and five segment rings in the hanging wall and six segment rings in the footwall also need to be reinforced.This study could provide a reference for metro design in ground fissure sites.展开更多
It has been argued that the human brain,as an information-processing machine,operates near a phase transition point in a non-equilibrium state,where it violates detailed balance leading to entropy production.Thus,the ...It has been argued that the human brain,as an information-processing machine,operates near a phase transition point in a non-equilibrium state,where it violates detailed balance leading to entropy production.Thus,the assessment of irreversibility in brain networks can provide valuable insights into their non-equilibrium properties.In this study,we utilized an open-source whole-brain functional magnetic resonance imaging(fMRI)dataset from both resting and task states to evaluate the irreversibility of large-scale human brain networks.Our analysis revealed that the brain networks exhibited significant irreversibility,violating detailed balance,and generating entropy.Notably,both physical and cognitive tasks increased the extent of this violation compared to the resting state.Regardless of the state(rest or task),interactions between pairs of brain regions were the primary contributors to this irreversibility.Moreover,we observed that as global synchrony increased within brain networks,so did irreversibility.The first derivative of irreversibility with respect to synchronization peaked near the phase transition point,characterized by the moderate mean synchronization and maximized synchronization entropy of blood oxygenation level-dependent(BOLD)signals.These findings deepen our understanding of the non-equilibrium dynamics of large-scale brain networks,particularly in relation to their phase transition behaviors,and may have potential clinical applications for brain disorders.展开更多
基金Project supported by the National Natural Science Foundation of China （Grant Nos 60374057 and 50575204）.
文摘In software engineering, class diagrams are often used to describe the system＇s class structures in Unified Modelling Language （UML）. A class diagram, as a graph, is a collection of static declarative model elements, such as classes, interfaces, and the relationships of their connections with each other. In this paper, class graphs axe examined within several Java software systems provided by Sun and IBM, and some new features are found. For a large-scale Java software system, its in-degree distribution tends to an exponential distribution, while its out-degree and degree distributions reveal the power-law behaviour. And then a directed preferential-random model is established to describe the corresponding degree distribution features and evolve large-scale Java software systems.
文摘To search for the Design Patterns’ influence on the software, the paper abstracts the feature models of 9 kinds of classic exiting design patterns among the 23 kinds and describes the features with algorithm language. Meanwhile, searching for the specific structure features in the network, the paper designs 9 matching algorithms of the 9 kinds design patterns mentioned above to research on the structure of the design patterns in the software network. At last, the paper analyzes the evolving trends of the software scale and the application frequency of the 9 kinds of design patterns as the software evolves, and search for the rules how these design patterns are applied into 4 kinds of typical software.
文摘Large-scale software systems,which are the most sophisticated human-designed objects,play more and more important role in our daily life.Consequently effective analysis for large-scale software has become an urgent problem to be solved with the increasing issues of software security and the continuous expansion of software applications scope.For the characteristics of large scale and complex structure in large-scale software,the traditional software analysis techniques are difficult to be used.With the problem of difficulty in presentation,storage and low efficiency in the process of large-scale software analysis,the visualization analysis framework for large-scale software based on software network,named SoNet,is proposed with the combination of complex network theory and program slicing technique.Constraint logic attributes of the programs will be obtained through source code parsing.Then we will construct a global view by the theory of complex network after extracting software structure and behavior,improving user’s perception of software architecture in a macro perspective.Use case slicing will be realized combined with Redis cluster,and accessibility analysis when given a keyword to be analyzed.We evaluate our prototype implementation on an open source software project named SoundSea in Github,and the results suggest that our approach can realize the analysis for large-scale software.
基金National Key R&D Program of China(Nos.2023YFC3806900,2022YFE0141400)。
文摘The advent of parametric design has resulted in a marked increase in the complexity of building.Unfortunately,traditional construction methods make it difficult to meet the needs.Therefore,construction robots have become a pivotal production tool in this context.Since the arm span of a single robot usually does not exceed 3 meters,it is not competent for producing large-scale building components.Accordingly,the extension of the robot,s working range is often achieved by external axes.Nevertheless,the coupling control of external axes and robots and their kinematic solution have become key challenges.The primary technical difficulties include customized construction robots,automatic solutions for external axes,fixed axis joints,and specific motion mode control.This paper proposes solutions to these difficulties,introduces the relevant basic concepts and algorithms in detail,and encapsulates these robotics principles and algorithm processes into the Grasshopper plug-in commonly used by architects to form the FURobot software platform.This platform effectively solves the above problems,lowers the threshold for architects,and improves production efficiency.The effectiveness of the algorithm and software in this paper is verified through simulation experiments.
基金supported by the National Key Research and Development Program of China(2022YFE0206700)。
文摘1.Introduction Climate change mitigation pathways aimed at limiting global anthropogenic carbon dioxide(CO_(2))emissions while striving to constrain the global temperature increase to below 2℃—as outlined by the Intergovernmental Panel on Climate Change(IPCC)—consistently predict the widespread implementation of CO_(2)geological storage on a global scale.
文摘Software security poses substantial risks to our society because software has become part of our life. Numerous techniques have been proposed to resolve or mitigate the impact of software security issues. Among them, software testing and analysis are two of the critical methods, which significantly benefit from the advancements in deep learning technologies. Due to the successful use of deep learning in software security, recently,researchers have explored the potential of using large language models(LLMs) in this area. In this paper, we systematically review the results focusing on LLMs in software security. We analyze the topics of fuzzing, unit test, program repair, bug reproduction, data-driven bug detection, and bug triage. We deconstruct these techniques into several stages and analyze how LLMs can be used in the stages. We also discuss the future directions of using LLMs in software security, including the future directions for the existing use of LLMs and extensions from conventional deep learning research.
基金funded by the Youth Fund of the National Natural Science Foundation of China(Grant No.42261070).
文摘Spectrum-based fault localization (SBFL) generates a ranked list of suspicious elements by using the program execution spectrum, but the excessive number of elements ranked in parallel results in low localization accuracy. Most researchers consider intra-class dependencies to improve localization accuracy. However, some studies show that inter-class method call type faults account for more than 20%, which means such methods still have certain limitations. To solve the above problems, this paper proposes a two-phase software fault localization based on relational graph convolutional neural networks (Two-RGCNFL). Firstly, in Phase 1, the method call dependence graph (MCDG) of the program is constructed, the intra-class and inter-class dependencies in MCDG are extracted by using the relational graph convolutional neural network, and the classifier is used to identify the faulty methods. Then, the GraphSMOTE algorithm is improved to alleviate the impact of class imbalance on classification accuracy. Aiming at the problem of parallel ranking of element suspicious values in traditional SBFL technology, in Phase 2, Doc2Vec is used to learn static features, while spectrum information serves as dynamic features. A RankNet model based on siamese multi-layer perceptron is constructed to score and rank statements in the faulty method. This work conducts experiments on 5 real projects of Defects4J benchmark. Experimental results show that, compared with the traditional SBFL technique and two baseline methods, our approach improves the Top-1 accuracy by 262.86%, 29.59% and 53.01%, respectively, which verifies the effectiveness of Two-RGCNFL. Furthermore, this work verifies the importance of inter-class dependencies through ablation experiments.
文摘Software-defined networking(SDN)is an innovative paradigm that separates the control and data planes,introducing centralized network control.SDN is increasingly being adopted by Carrier Grade networks,offering enhanced networkmanagement capabilities than those of traditional networks.However,because SDN is designed to ensure high-level service availability,it faces additional challenges.One of themost critical challenges is ensuring efficient detection and recovery from link failures in the data plane.Such failures can significantly impact network performance and lead to service outages,making resiliency a key concern for the effective adoption of SDN.Since the recovery process is intrinsically dependent on timely failure detection,this research surveys and analyzes the current literature on both failure detection and recovery approaches in SDN.The survey provides a critical comparison of existing failure detection techniques,highlighting their advantages and disadvantages.Additionally,it examines the current failure recovery methods,categorized as either restoration-based or protection-based,and offers a comprehensive comparison of their strengths and limitations.Lastly,future research challenges and directions are discussed to address the shortcomings of existing failure recovery methods.
文摘Software-related security aspects are a growing and legitimate concern,especially with 5G data available just at our palms.To conduct research in this field,periodic comparative analysis is needed with the new techniques coming up rapidly.The purpose of this study is to review the recent developments in the field of security integration in the software development lifecycle(SDLC)by analyzing the articles published in the last two decades and to propose a way forward.This review follows Kitchenham’s review protocol.The review has been divided into three main stages including planning,execution,and analysis.From the selected 100 articles,it becomes evident that need of a collaborative approach is necessary for addressing critical software security risks(CSSRs)through effective risk management/estimation techniques.Quantifying risks using a numeric scale enables a comprehensive understanding of their severity,facilitating focused resource allocation and mitigation efforts.Through a comprehensive understanding of potential vulnerabilities and proactive mitigation efforts facilitated by protection poker,organizations can prioritize resources effectively to ensure the successful outcome of projects and initiatives in today’s dynamic threat landscape.The review reveals that threat analysis and security testing are needed to develop automated tools for the future.Accurate estimation of effort required to prioritize potential security risks is a big challenge in software security.The accuracy of effort estimation can be further improved by exploring new techniques,particularly those involving deep learning.It is also imperative to validate these effort estimation methods to ensure all potential security threats are addressed.Another challenge is selecting the right model for each specific security threat.To achieve a comprehensive evaluation,researchers should use well-known benchmark checklists.
文摘Link failure is a critical issue in large networks and must be effectively addressed.In software-defined networks(SDN),link failure recovery schemes can be categorized into proactive and reactive approaches.Reactive schemes have longer recovery times while proactive schemes provide faster recovery but overwhelm the memory of switches by flow entries.As SDN adoption grows,ensuring efficient recovery from link failures in the data plane becomes crucial.In particular,data center networks(DCNs)demand rapid recovery times and efficient resource utilization to meet carrier-grade requirements.This paper proposes an efficient Decentralized Failure Recovery(DFR)model for SDNs,meeting recovery time requirements and optimizing switch memory resource consumption.The DFR model enables switches to autonomously reroute traffic upon link failures without involving the controller,achieving fast recovery times while minimizing memory usage.DFR employs the Fast Failover Group in the OpenFlow standard for local recovery without requiring controller communication and utilizes the k-shortest path algorithm to proactively install backup paths,allowing immediate local recovery without controller intervention and enhancing overall network stability and scalability.DFR employs flow entry aggregation techniques to reduce switch memory usage.Instead of matching flow entries to the destination host’s MAC address,DFR matches packets to the destination switch’s MAC address.This reduces the switches’Ternary Content-Addressable Memory(TCAM)consumption.Additionally,DFR modifies Address Resolution Protocol(ARP)replies to provide source hosts with the destination switch’s MAC address,facilitating flow entry aggregation without affecting normal network operations.The performance of DFR is evaluated through the network emulator Mininet 2.3.1 and Ryu 3.1 as SDN controller.For different number of active flows,number of hosts per edge switch,and different network sizes,the proposed model outperformed various failure recovery models:restoration-based,protection by flow entries,protection by group entries and protection by Vlan-tagging model in terms of recovery time,switch memory consumption and controller overhead which represented the number of flow entry updates to recover from the failure.Experimental results demonstrate that DFR achieves recovery times under 20 milliseconds,satisfying carrier-grade requirements for rapid failure recovery.Additionally,DFR reduces switch memory usage by up to 95%compared to traditional protection methods and minimizes controller load by eliminating the need for controller intervention during failure recovery.Theresults underscore the efficiency and scalability of the DFR model,making it a practical solution for enhancing network resilience in SDN environments.
基金National Natural Science Foundation of China(52375378)National Key Laboratory of Metal Forming Technology and Heavy Equipment(S2308100.W12)Huxiang High-Level Talent Gathering Project of Hunan Province(2021RC5001)。
文摘The titanium alloy strut serves as a key load-bearing component of aircraft landing gear,typically manufactured via forging.The friction condition has important influence on material flow and cavity filling during the forging process.Using the previously optimized shape and initial position of preform,the influence of the friction condition(friction factor m=0.1–0.3)on material flow and cavity filling was studied by numerical method with a shear friction model.A novel filling index was defined to reflect material flow into left and right flashes and zoom in on friction-induced results.The results indicate that the workpiece moves rigidly to the right direction,with the displacement decreasing as m increases.When m<0.18,the underfilling defect will occur in the left side of strut forging,while overflow occurs in the right forging die cavity.By combining the filling index and analyses of material flow and filling status,a reasonable friction factor interval of m=0.21–0.24 can be determined.Within this interval,the cavity filling behavior demonstrates robustness,with friction fluctuations exerting minimal influence.
基金National Natural Science Foundation of China(12471367)。
文摘In this work,we present a parallel implementation of radiation hydrodynamics coupled with particle transport,utilizing software infrastructure JASMIN(J Adaptive Structured Meshes applications INfrastructure)which encapsulates high-performance technology for the numerical simulation of complex applications.Two serial codes,radiation hydrodynamics RH2D and particle transport Sn2D,have been integrated into RHSn2D on JASMIN infrastructure,which can efficiently use thousands of processors to simulate the complex multi-physics phenomena.Moreover,the non-conforming processors strategy has ensured RHSn2D against the serious load imbalance between radiation hydrodynamics and particle transport for large scale parallel simulations.Numerical results show that RHSn2D achieves a parallel efficiency of 17.1%using 90720 cells on 8192 processors compared with 256 processors in the same problem.
文摘Quantum software development utilizes quantum phenomena such as superposition and entanglement to address problems that are challenging for classical systems.However,it must also adhere to critical quantum constraints,notably the no-cloning theorem,which prohibits the exact duplication of unknown quantum states and has profound implications for cryptography,secure communication,and error correction.While existing quantum circuit representations implicitly honor such constraints,they lack formal mechanisms for early-stage verification in software design.Addressing this constraint at the design phase is essential to ensure the correctness and reliability of quantum software.This paper presents a formal metamodeling framework using UML-style notation and and Object Constraint Language(OCL)to systematically capture and enforce the no-cloning theorem within quantum software models.The proposed metamodel formalizes key quantum concepts—such as entanglement and teleportation—and encodes enforceable invariants that reflect core quantum mechanical laws.The framework’s effectiveness is validated by analyzing two critical edge cases—conditional copying with CNOT gates and quantum teleportation—through instance model evaluations.These cases demonstrate that the metamodel can capture nuanced scenarios that are often mistaken as violations of the no-cloning theorem but are proven compliant under formal analysis.Thus,these serve as constructive validations that demonstrate the metamodel’s expressiveness and correctness in representing operations that may appear to challenge the no-cloning theorem but,upon rigorous analysis,are shown to comply with it.The approach supports early detection of conceptual design errors,promoting correctness prior to implementation.The framework’s extensibility is also demonstrated by modeling projective measurement,further reinforcing its applicability to broader quantum software engineering tasks.By integrating the rigor of metamodeling with fundamental quantum mechanical principles,this work provides a structured,model-driven approach that enables traditional software engineers to address quantum computing challenges.It offers practical insights into embedding quantum correctness at the modeling level and advances the development of reliable,error-resilient quantum software systems.
基金supported by the CCF-NSFOCUS‘Kunpeng’Research Fund(CCF-NSFOCUS2024012).
文摘In recent years,with the rapid development of software systems,the continuous expansion of software scale and the increasing complexity of systems have led to the emergence of a growing number of software metrics.Defect prediction methods based on software metric elements highly rely on software metric data.However,redundant software metric data is not conducive to efficient defect prediction,posing severe challenges to current software defect prediction tasks.To address these issues,this paper focuses on the rational clustering of software metric data.Firstly,multiple software projects are evaluated to determine the preset number of clusters for software metrics,and various clustering methods are employed to cluster the metric elements.Subsequently,a co-occurrence matrix is designed to comprehensively quantify the number of times that metrics appear in the same category.Based on the comprehensive results,the software metric data are divided into two semantic views containing different metrics,thereby analyzing the semantic information behind the software metrics.On this basis,this paper also conducts an in-depth analysis of the impact of different semantic view of metrics on defect prediction results,as well as the performance of various classification models under these semantic views.Experiments show that the joint use of the two semantic views can significantly improve the performance of models in software defect prediction,providing a new understanding and approach at the semantic view level for defect prediction research based on software metrics.
基金supported in part by the Teaching Reform Project of Chongqing University of Posts and Telecommunications,China under Grant No.XJG23234Chongqing Municipal Higher Education Teaching Reform Research Project under Grant No.203399the Doctoral Direct Train Project of Chongqing Science and Technology Bureau under Grant No.CSTB2022BSXM-JSX0007。
文摘The advent of large language models(LLMs)has made knowledge acquisition and content creation increasingly easier and cheaper,which in turn redefines learning and urges transformation in software engineering education.To do so,there is a need to understand the impact of LLMs on software engineering education.In this paper,we conducted a preliminary case study on three software requirements engineering classes where students are allowed to use LLMs to assist in their projects.Based on the students’experience,performance,and feedback from a survey conducted at the end of the courses,we characterized the challenges and benefits of applying LLMs in software engineering education.This research contributes to the ongoing discourse on the integration of LLMs in education,emphasizing both their prominent potential and the need for balanced,mindful usage.
文摘This paper presents our endeavors in developing the large-scale, ultra-high-resolution E3SM Land Model (uELM), specifically designed for exascale computers furnished with accelerators such as Nvidia GPUs. The uELM is a sophisticated code that substantially relies on High-Performance Computing (HPC) environments, necessitating particular machine and software configurations. To facilitate community-based uELM developments employing GPUs, we have created a portable, standalone software environment preconfigured with uELM input datasets, simulation cases, and source code. This environment, utilizing Docker, encompasses all essential code, libraries, and system software for uELM development on GPUs. It also features a functional unit test framework and an offline model testbed for comprehensive numerical experiments. From a technical perspective, the paper discusses GPU-ready container generations, uELM code management, and input data distribution across computational platforms. Lastly, the paper demonstrates the use of environment for functional unit testing, end-to-end simulation on CPUs and GPUs, and collaborative code development.
基金The Australian Research Council(DP200101197,DP230101107).
文摘Formalizing complex processes and phenomena of a real-world problem may require a large number of variables and constraints,resulting in what is termed a large-scale optimization problem.Nowadays,such large-scale optimization problems are solved using computing machines,leading to an enormous computational time being required,which may delay deriving timely solutions.Decomposition methods,which partition a large-scale optimization problem into lower-dimensional subproblems,represent a key approach to addressing time-efficiency issues.There has been significant progress in both applied mathematics and emerging artificial intelligence approaches on this front.This work aims at providing an overview of the decomposition methods from both the mathematics and computer science points of view.We also remark on the state-of-the-art developments and recent applications of the decomposition methods,and discuss the future research and development perspectives.
文摘This article focuses on the management of large-scale machinery and equipment in highway construction,with the research objective of identifying issues at the management level and exploring more effective management measures.Through practical observation and logical analysis,this article elaborates on the management connotations of large-scale machinery and equipment in highway construction,affirming its management value from different perspectives.On this basis,it carefully analyzes the problems existing in the management of large-scale machinery and equipment,providing a detailed interpretation of issues such as the weak foundation of the equipment management system and the disconnection between equipment selection and configuration from reality.Combining the manifestations of related problems,this article proposes strategies such as strengthening the institutional foundation of equipment management,selecting and configuring equipment based on actual conditions,aiming to provide references for large-scale machinery and equipment management to relevant enterprises.
基金supported by the National Key Research&Development Program of China(Grant No.2023YFC3008404)the Key Laboratory of Earth Fissures Geological Disaster,Ministry of Natural Resources,China(Grant Nos.EFGD20240609 and EFGD20240610).
文摘The recent upsurge in metro construction emphasizes the necessity of understanding the mechanical performance of metro shield tunnel subjected to the influence of ground fissures.In this study,a largescale experiment,in combination with numerical simulation,was conducted to investigate the influence of ground fissures on a metro shield tunnel.The results indicate that the lining contact pressure at the vault increases in the hanging wall while decreases in the footwall,resulting in a two-dimensional stress state of vertical shear and axial tension-compression,and simultaneous vertical dislocation and axial tilt for the segments around the ground fissure.In addition,the damage to curved bolts includes tensile yield,flexural yield,and shear twist,leading to obvious concrete lining damage,particularly at the vault,arch bottom,and hance,indicating that the joints in these positions are weak areas.The shield tunnel orthogonal to the ground fissure ultimately experiences shear failure,suggesting that the maximum actual dislocation of ground fissure that the structure can withstand is approximately 20 cm,and five segment rings in the hanging wall and six segment rings in the footwall also need to be reinforced.This study could provide a reference for metro design in ground fissure sites.
基金supported by the Fundamental Research Funds for the Central Universities(Grant Nos.lzujbky-2021-62 and lzujbky-2024-jdzx06)the National Natural Science Foundation of China(Grant No.12247101)+1 种基金the Natural Science Foundation of Gansu Province,China(Grant Nos.22JR5RA389 and 23JRRA1740)the‘111 Center’Fund(Grant No.B20063).
文摘It has been argued that the human brain,as an information-processing machine,operates near a phase transition point in a non-equilibrium state,where it violates detailed balance leading to entropy production.Thus,the assessment of irreversibility in brain networks can provide valuable insights into their non-equilibrium properties.In this study,we utilized an open-source whole-brain functional magnetic resonance imaging(fMRI)dataset from both resting and task states to evaluate the irreversibility of large-scale human brain networks.Our analysis revealed that the brain networks exhibited significant irreversibility,violating detailed balance,and generating entropy.Notably,both physical and cognitive tasks increased the extent of this violation compared to the resting state.Regardless of the state(rest or task),interactions between pairs of brain regions were the primary contributors to this irreversibility.Moreover,we observed that as global synchrony increased within brain networks,so did irreversibility.The first derivative of irreversibility with respect to synchronization peaked near the phase transition point,characterized by the moderate mean synchronization and maximized synchronization entropy of blood oxygenation level-dependent(BOLD)signals.These findings deepen our understanding of the non-equilibrium dynamics of large-scale brain networks,particularly in relation to their phase transition behaviors,and may have potential clinical applications for brain disorders.
