In high-dimensional data processing,the K-nearest neighbor(KNN)graph is a critical data structure widely used in tasks such as clustering,graph neural networks,and recommendation systems.However,with the increasing use of pretrained embedding models in unstructured data modeling and retrieval,embedding model fine-tuning has become a key step in enhancing the semantic representation capability of embeddings.Such fine-tuning often leads to systematic changes in the vector representations of all data points,which invalidates the original neighborhood relationships in the KNN graph.Existing research primarily focuses on building KNN graphs for static data,lacking efficient solutions for adapting to updated embeddings after fine-tuning.To address this gap,this study proposes FastAdjust,an efficient KNN graph update method tailored for embedding model fine-tuning scenarios.Leveraging the observation that fine-tuning usually causes only minor changes to individual embeddings,incremental adjustments to the original KNN graph are performed by FastAdjust through a local update strategy,significantly improving update efficiency while maintaining graph quality.Specifically,FastAdjust first employs a clustering structure based on product quantization to efficiently and accurately locate a subset of candidate neighbors for each data point,thus narrowing the search space.Secondly,based on data density and the magnitude of embedding variation,FastAdjust leverages their correlation with changes in the KNNs to adaptively allocate update resources according to the degree of neighbor relationship changes,thus improving overall update efficiency.Experimental results on real-world datasets demonstrate that FastAdjust efficiently and accurately adapts KNN graphs to embedding updates with significantly reduced computational cost,showing strong practical value and scalability.
With the widespread application of multi-source,heterogeneous,and multi-modal data in scenarios such as large models and data lakes,there has been a significant growth in vector-based data retrieval and storage management.By mapping heterogeneous data into high-dimensional vector representations and leveraging vector indices,vector databases facilitate the unified management of diverse data types and enable high-quality similarity search,establishing them as a crucial foundation for applications like generative retrieval and AI-native databases.However,existing vector databases face significant bottlenecks in terms of storage and indexing efficiency,index construction complexity,and retrieval accuracy.Specifically,massive high-dimensional vectors lead to increased storage overhead and maintenance costs for indices.Furthermore,vector index structures are often bloated,resulting in substantial memory consumption.Moreover,the degradation of retrieval accuracy caused by distortion from compression techniques remains an unresolved challenge.This study proposes a framework based on weight residual vector quantization(WRVQ).This method achieves efficient compression and storage with very low distortion by decoupling the quantization direction from the residual magnitude.It stores the residual direction as a unit vector and appends a weight marker.For indexing,a three-layer inverted index structure tailored to the characteristics of WRVQ is designed,comprising an exact match layer,a fuzzy match layer,and a search layer.This structure organically integrates asymmetric distance computation(ADC)with nearest neighbor search techniques to realize approximate nearest neighbor(ANN)search that balances both high accuracy and high efficiency.Experimental results on large-scale datasets demonstrate that,compared to traditional low-dimensional embedding models and existing quantization methods,WRVQ achieves significant improvements across key metrics,including quantization loss,storage compression ratio,and retrieval recall.Furthermore,it exhibits considerable advantages in both index construction and query performance.
Text2SQL has evolved into a significant tool for data analysis and database management by reducing the technical barriers for non-expert users to interact with relational databases.The introduction of large language model(LLM),represented by GPT,further improves the performance of Text2SQL systems.However,existing Text2SQL techniques are difficult to apply directly to the spatial database domain because spatial data involves complex geometric relationships,diverse query types,and the demand for high-precision semantic understanding.To address these issues and lower the threshold for interaction between non-experts and spatial databases,a natural language query(NLQ)transformation method for spatial databases is proposed.The method consists of two core phases:(1)natural language understanding;(2)executable language generation.In phase(1),an entity information extraction algorithm is employed to extract key query entities,and a spatial data query corpus is constructed based on large language models to determine the query type.In phase(2),a structured language model(SLM)is selected according to the query type,and the entities are then mapped into the structured language model to generate the final executable language for spatial databases.Experimental results on multiple real-world datasets demonstrate that the proposed method enables efficient transformation from natural language queries to executable languages of spatial databases.
In certain designs and applications of practical lattice-based cryptography,the use of a specialized variant of LWE problems,where the public matrix is sampled from a non-uniform distribution,is required to establish the securities of corresponding cryptographic schemes.Recently,the formal definition of LWE problems with semi-uniform seeds was introduced in some work,in which the hardness of Euclidean,ideal,and module lattice-based LWE problems with semi-uniform seeds was proved through reduction roadmaps similar to those employed in the hardness proofs of entropic LWE problems.However,known reduction introduces significant losses in the Gaussian parameters of errors and dimensions.Moreover,additional non-standard assumptions are required to demonstrate the hardness of LWE problems with semi-uniform seeds over rings.In this study,a tighter reduction is proposed for LWE problems with semi-uniform seeds by incorporating modified techniques from the hardness proofs of Hint-LWE problems.The proposed reduction is largely unaffected by the algebraic structures of the underlying problems and can be uniformly applied to Euclidean,ideal,and module lattice-based LWE problems with semi-uniform seeds.The hardness of these LWE problems can be established based on standard LWE assumptions without the need for any additional non-standard assumptions.Furthermore,the dimension of the corresponding LWE problems remains unchanged,and the reduction introduces only minimal losses in Gaussian parameters of errors.
Time-series data are widely used in fields such as industrial manufacturing,meteorology,electric power,and vehicles,which has spurred the development of time-series database management systems.More and more database systems are migrating to the cloud,and the architecture of end-cloud collaboration is becoming more common,leading to increasingly large data scales to be processed.In scenarios such as end-cloud collaboration and massive time series,a large number of short time series are generated due to short synchronization cycles and frequent data flushing,among other reasons,presenting new challenges to database systems.Efficient data management and compression methods can significantly improve storage performance,enabling database systems to handle the storage of massive time series.Apache TsFile is a columnar storage file format specifically designed for time series scenarios,playing an important role in database management systems such as Apache IoTDB.This study elaborates on the group compression and merging methods used in Apache TsFile to address scenarios with a large number of short time series,especially in application scenarios with a vast number of time series such as the Industrial Internet of Things.This group compression method fully considers the data characteristics in the short time series scenario.Through device grouping,it improves metadata utilization,reduces file index size,decreases short time series,and significantly improves compression effectiveness.After validation with real-world datasets,the proposed grouping method shows significant improvements in compression effect,reading,writing,file merging,and other aspects,enabling better management of TsFiles in short time series scenarios.
Databases are important foundational components in computer services.However,performance anomalies may occur during their operation,affecting business service quality.How to diagnose performance anomalies in databases has become a hot issue in industry and academia.Recently,a series of automated database anomaly diagnosis methods have been successively proposed.They analyze the runtime status of the database and determine the overall database anomaly types.However,with the continuous expansion of data scale,distributed databases are becoming an increasingly popular solution in the industry.In a distributed database,which is composed of multiple nodes,existing anomaly diagnosis methods struggle to effectively locate node anomalies,fail to identify compound anomalies across multiple nodes,and are unable to perceive the complex performance influence relationships between nodes,lacking effective diagnostic capabilities.To address these challenges,this study proposes a distributed database diagnosis method for compound anomalies,named DistDiagnosis.It models the anomalous state of distributed databases using a Compound Anomaly Graph,which not only represents anomalies at each node but also effectively captures the correlations between nodes.DistDiagnosis introduces a node correlation-aware root cause anomaly ranking method,effectively locating root cause anomalies according to the influence of nodes on the database.In this study,anomaly testing cases for various scenarios are constructed on OceanBase,a domestically developed distributed database.Experimental results show that DistDiagnosis outperforms other advanced baselines,achieving the AC@1,AC@3,and AC@5 values of 0.97,0.98,and 0.98.Compared to the second-best method,DistDiagnosis improves accuracy by up to 5.20%,5.45%,and 4.46%in each diagnostic scenario.
Single-master multi-slave is the mainstream architecture of cloud-native databases.In the cluster,slave nodes can share the read-only requests of the master node,while write requests are handled by the master node.Based on this,to further meet the demands of large-scale transaction expansion,some cloud databases attempt to implement multi-write transaction expansion.One possible approach to multi-write expansion is to introduce shared cache among computing nodes to support cross-node data access.For shared-cache database systems,the overhead of cross-node remote access is significantly higher than that of local access.Therefore,the design of cache protocol is a crucial factor that affects system performance and scalability.This study proposes two innovative improvements to the coherence protocol and implements PG-RAC,a shared-cache database,which supports multi-write transactions based on PostgreSQL.On one hand,PG-RAC proposes a new distributed chained routing strategy,which disperses routing information among computing nodes.Compared to the routing strategy that utilizes single-node directory management,it reduces the average transaction latency by approximately 20%.On the other hand,this study also enhances the duplicate page invalidation mechanism by separating invalidation operations from the transaction path,reducing the latency of the critical path in the transaction.Based on this,PG-RAC takes advantage of the characteristics of multi-version concurrency control(MVCC)and further proposes to delay the invalidation point of duplicate pages,which effectively improves cache utilization.TPC-C experimental results show that for a cluster with 4 compute nodes,the throughput is nearly 2 times that of PostgreSQL and 1.5 times that of the distributed database Citus.
Cloud-native databases,with advantages such as out-of-the-box functionality,elastic scalability,and pay-as-you-go,are currently a research hotspot in academia and industry.Currently,cloud-native databases only support"single writer and multiple readers",that is,read-write transactions are concentrated on a single read-write node,and read-only transactions are distributed to multiple read-only nodes.This limitation restricts the system's ability to process read-write transactions,making it difficult to meet the demands of write-intensive businesses.To this end,this study proposes the D3C(deterministic concurrency control cloud-native database)architecture.It breaks through the limitation of"single writer and multiple readers"and supports concurrency execution of read-write transactions on multiple read-write nodes by designing a cloud-native database transaction processing mechanism based on deterministic concurrency control.D3C splits transactions into sub-transactions and independently executes them on each node according to a predefined global order,ensuring serializability for transaction execution on multiple read-write nodes.Additionally,this study introduces mechanisms like asynchronous batch data persistence mechanisms based on multi-version to ensure transaction processing performance and proposes a consistency point-based fault recovery mechanism to achieve high availability.Experimental results show that D3C can achieve 5.1 times the performance of the"single writer and multiple readers"architecture in write-intensive scenarios while meeting the key requirements of cloud-native databases.
One of the most important features of multi-tenant databases in cloud environments is scalability.However,most elastic scaling techniques struggle to make effective scaling decisions for dynamically changing loads.If load changes can be predicted in advance,resource supply can be accurately adjusted.Given this,this study proposes a load-prediction-based elastic scaling method for multi-tenant databases.It includes a combined load prediction model and an elastic scaling strategy.The load prediction model combines the advantages of convolutional neural networks,long short-term memory networks and gated recurrent units.It can accurately forecast memory requirements of database clusters.Based on the prediction results,the elastic scaling strategy adjusts the number of virtual machines to ensure that resource supply remains within a reasonable range.Compared to existing methods,the combined load prediction model can reduce prediction errors by 8.7%to 21.8%and improve prediction fitting degree by 4.6%.Furthermore,this study improves the Bayesian optimization algorithm for hyperparameter tuning of the combined prediction model.The improved hyperparameter tuning model reduces errors by above 20%and improves fitting degree by 1.04%,which proves that it can well address the poor performance of Bayesian optimization in combined domains of discrete and continuous solutions.Compared to the most widely used scaling strategy in Kubernetes,the proposed elastic scaling method reduces response time by 8.12%and latency by 9.56%.It can avoid the latency and the waste of resources to a large extent.
As a recursive method for finding the optimal solution to a problem,dynamic programming mainly solves the original problem by first solving the subproblems and then combining their solutions.Due to a large number of dependencies and constraints among its subproblems,the validation procedure is laborious,and especially the correctness verification of imperative dynamic programming algorithms is a challenge.Based on the functional modeling and verification of dynamic programming algorithms Isabelle/HOL,this study avoids dealing with complex dependencies and constraints in proving correctness by verifying the equivalence of imperative dynamic programming algorithms and their programs.Meanwhile,a framework for the design of imperative dynamic programming algorithmic programs and their mechanized verification are proposed.First,according to the optimization method(memo method)and properties(optimal substructure property and subproblems overlapping property)of dynamic programming algorithms,the problem specification is described,the recursive relations are inductively derived,and the loop invariants are formally constructed.Then,the IMP(minimalistic imperative programming language)code is generated based on the recursive relations.Second,the problem specification,loop invariants,and generated IMP code are fed into VCG(verification condition generator)to generate the verification condition for correctness automatically.Additionally,the verification condition is then mechanically verified in the Isabelle/HOL theorem prover.The algorithm is initially designed in the general form of an imperative dynamic programming algorithm and further instantiated to obtain specific algorithms.Finally,the effectiveness of the proposed framework is validated by case studies to provide references for automated derivation and verification of dynamic programming algorithms.
Operating systems are the key foundational components of the software stacks employed in many safety-critical scenarios.A tiny error or loophole in the operating system may cause major failures of the overall software system,resulting in huge economic losses or endangering human lives.Thus,the correctness of the operating system should be verified to reduce the number of such accidents.Traditional testing methods cannot guarantee the exhaustive detection of potential errors in the target system.Therefore,it is necessary to adopt formal methods based on strict mathematical theories for verifying operating systems.In an operating system,mutexes are utilized to coordinate the access of shared resources by tasks and they are a typical means of task synchronization.The functional correctness of mutexes is the key to the correct functioning of multi-task applications.Based on the theorem proof method,this study conducts formal verification on the code of the mutex module of a preemptive microkernel in an interactive theorem prover Coq,gives the formal specifications of the interface functions of this module,and formally proves the functional correctness of these interface functions.
Linearizability is universally accepted as a correctness criterion for concurrent objects.However,it has been shown that linearizability cannot be adopted as a correctness criterion for concurrent objects with random sentences.Thus,Golab et al.proposed the concept of strong linearizability,which adds prefix preservation properties based on the linearizability definition and has more constraints for concurrent objects.The research on strong linearizability focuses on the feasibility of generating strongly linearizable objects with certain basic objects,while only a few studies are about checking and verification of strong linearizability.This study investigates strong linearizability from two aspects including the verification algorithm and approach for proving non-strong linearizability of concurrent objects.First,it divides strong linearizability into fixed effective points and pure help and proves that the notion of fixed effective points is an extension of that of fixed linearizability points.Then,two verification algorithms for strong linearizability are put forward.One algorithm is based on checking the fixed linearizability points,and the other is based on the fixed effective points.Finally,an approach is provided for proving that the concurrent objects violate strong linearizability,and it helps verify that the Herlihy&Wing queue,a single-reader single-write register,and a snapshot object violate strong linearizability.
Many complex embedded systems are mixed-criticality systems(MCSs).MCSs are often required to operate with the specified criticality level,but they may be subject to hazards that can induce random errors and burst errors,which may result in the abortion of an executing thread or even system failures.Current research only concentrates on schedulability analysis for MCSs and fails to further analyze system safety and consider the dependency relationship among threads.Taking random errors and burst errors as the research objects,this study proposes an architecture-based MCS safety analysis method with the integration of fault propagation analysis.Meanwhile,architecture analysis and design language(AADL)is employed to characterize the dependency relations among components.To compensate for the shortcomings of AADL,this study creates new AADL properties(AADL burst error properties)and proposes new thread state machine(burst error-based thread state machine)semantics to describe the thread execution process with burst errors.Additionally,model transformation rules and assembly methods are proposed to apply probabilistic model checking for safety analysis,and PRISM models are derived from AADL models.Two formulae are also formulated to obtain quantitative safety properties for verifying occurrence probabilities of failures,and qualitative safety properties for generating corresponding witnesses to figure out propagation paths for fault propagation analysis respectively.Finally,the effectiveness of the proposed method is verified by adopting a power boat autopilot(PBA)system.
Existing work on the analysis of real-time embedded multi-core systems using point-interval prioritized time Petri nets has the following limitations.(1)Point-interval prioritized time Petri nets only consider the case where the execution time of each task is a fixed value,but in many practical applications,the execution time of a task is generally within a range so that this kind of model cannot be used to analyze these applications.(2)There is a lack of automatic transformation from task dependency graphs to this point-interval prioritized time Petri nets,and thus it is inconvenient for engineering designers.(3)The case of mutually exclusive access to shared variables has not been considered.To address these issues,this study defines prioritized time Petri nets(Pri-TPN)to overcome the first limitation and introduces a task dependency graph with resource allocation and priority(TDG-RAP)to overcome the third limitation.It develops algorithms based on Pri-TPN for analyzing the worst-case execution time(WCET)and system deadlocks of tasks.Additionally,a tool software is developed to facilitate its use by engineering designers.
Deep learning-based code vulnerability detection models have gradually become an important method for detecting software vulnerabilities due to their advantages of high detection efficiency and accuracy,and play an important role in the code auditing service of the code hosting platform GitHub.However,deep neural networks have been proved to be susceptible to the interference of adversarial attacks,which leads to the risk of deep learning-based vulnerability detection models being attacked and reducing the detection accuracy.Therefore,building adversarial attacks against vulnerability detection models can not only uncover the security flaws of such models,but also help to evaluate the robustness of the models,and then improve the performance of the models through corresponding methods.However,the existing counter-attack methods for vulnerability detection models rely on generalized code transformation tools,and do not propose targeted code perturbation operations and decision algorithms,so it is difficult to generate effective counter-attack samples,and the legitimacy of the counter-attack samples relies on manual checking.To address the above problems,a reinforcement learning adversarial attack method for vulnerability detection model is proposed.The method firstly designs a series of semantically constrained and vulnerability-preserving code perturbation operations as a set of perturbations;secondly,the code samples with vulnerabilities are used as inputs,and the reinforcement learning model is used to select specific sequences of perturbation operations;finally,the code samples are used to search for potential locations of perturbations according to the types of nodes in the syntax tree,and then code transformations are carried out,thus generating the counteracting samples.Based on SARD and NVD,two experimental datasets with a total of 14 278 code samples are constructed,and four vulnerability detection models with different characteristics are trained as attack targets.For each target model,a reinforcement learning network is trained to counter the attack.The results show that the attack method leads to a 74.34%decrease in the recall of the models and a 96.71%success rate,which is an average increase of 68.76%compared to the baseline method.The experiment proves that the current vulnerability detection model has the risk of being attacked,and further research is needed to improve the robustness of the model.
In recent years,with the rise of the mobile Internet,underground mobile applications primarily involved in scams,gambling,and pornography have become more rampant,requiring effective control measures.Currently,there is a lack of research on underground applications by researchers.Due to the continuous crackdown by law enforcement agencies on traditional distribution channels for these applications,the existing collection methods based on search engines and app stores have proven to be ineffective.The lack of large-scale and representative datasets of real-world underground applications has become a major constraint for in-depth research.Therefore,this study aims to address the challenge of collection of large-scale real-world underground applications,providing data support for a comprehensive in-depth analysis of these applications and their ecosystem.A method is proposed to capture underground applications based on traffic analysis.By focusing on the key distribution channels of underground applications and leveraging their characteristics of mutation and accompanying traffic,underground applications can be discovered in the propagation stage.In the test,the proposed method successfully obtained 3 439 application download links and 3 303 distinct applications.Among these apps,91.61%of the samples were labeled as malware by antivirus engine,while 98.14%of the samples were zero-days.The results demonstrate the effectiveness of the proposed method in the collection of underground applications.
Through providing a virtual environment modeled from embedded devices,firmware rehosting enables dynamic analysis on embedded device firmware.Existing full-emulation firmware hosting solutions can only preventatively fix known hardware and software dependencies but cannot address undetected dependencies during the rehosting process.This study proposes FirmDep,an embedded application rehosting solution assisted with dynamic analysis.During the rehosting process,FirmDep records the execution trace and system state of the embedded application to be analyzed.If FirmDep fails to rehost the application,FirmDep extracts information and recover system states from the execution trace,then uses several algorithms to identify and arbitrate the unresolved dependency problems.The prototype system of FirmDep is implemented based on PANDA and angr,and it is tested with embedded Web applications from 217 real-world firmware images.The results show that FirmDep can effectively identify unresolved dependencies of embedded application and improve the success rate of rehosting.
The privacy policy statement of a mobile application serves as a crucial document that must be disclosed to users before collecting their information.However,current privacy policy statements face various issues,such as missing key disclosure items,omitting information collection purposes,and using vague descriptions.With an increasing number of legal provisions,the requirements for privacy policy statements vary,making compliance verification more burdensome.This study proposes a multi-label classification method for mobile application privacy policy statements.This method compares the requirements of four core laws and regulations regarding privacy policy statements,summarizes and organizes 31 categories of core item labels and features.Under this label system,the study designs and implements a classification model for privacy policy statement sentences,which achieves a 94%accuracy rate in item classification.Using this model,compliance verification was conducted in Android applications and mini-program scenarios,revealing issues such as missing items(79%),omitted purposes(63%),and vague descriptions(94%)in privacy policy statements.
Graph data is a kind of data composed of nodes and edges,which models the entities as the nodes,nodes may be connected by edges,and edge indicates a relationship between entities.By analyzing and mining these data,a lot of valuable information can be revealed.Meanwhile,it also brings risks of privacy information disclosure for every entity in the graph.To address this issue,a graph data publishing method is proposed based on the negative database(NDB).This method transforms the structural characteristics of the graph data into the encoding format of a negative database.Based on this,a generation method for perturbed graphs(NDB-Graph)is designed.Since NDB is a privacy-preserving technique that does not explicitly store the original data and is difficult to reverse,the published graph data ensures the security of the original graph data.Besides,due to the high efficiency of graph neural network in relation feature processing in graph data,it is widely used in various task processing modeling on graph data,such as recommendation system.a graph neural network recommendation system is also proposed based on NDB technology to protect the privacy of graph data for each user.Compared with publishing method PBCN,the proposed method outperforms it in most cases in experiments on the Karate and Facebook datasets.For example,on Facebook datasets,the smallest L1-error of degree distribution is only 6,which is about 2.6%lower than the PBCN method under the same privacy level,the worst case is about 1 400,which is about 46.5%lower than the PBCN method under the same privacy level.The experiment of collaborative filtering based on LightGCN also demonstrates that the proposed privacy protection method has high precision.
With the advent of the big data era,the significance of data analysis has increasingly come to the forefront,showcasing its ability to uncover valuable insights from vast datasets,thereby enhancing the decision-making process for users.Nonetheless,the data analysis workflow faces three dominant challenges:high coupling in the analysis workflow,a plethora of interactive interfaces,and a time-intensive exploratory analysis process.To address these challenges,this study introduces Navi,a data analysis system powered by natural language interaction.Navi embraces a modular design philosophy that abstracts three core functional modules from mainstream data analysis workflows:data querying,visualization generation,and visualization exploration.This approach effectively reduces the coupling of the system.Meanwhile,Navi leverages natural language as a unified interactive interface to seamlessly integrate various functional modules through a task scheduler,ensuring their effective collaboration.Moreover,in order to address the challenges of exponential search space and ambiguous user intent in visualization exploration,this study proposes an automated approach for visualization exploration based on Monte Carlo tree search.In addition,a pruning algorithm and a composite reward function,both incorporating visualization domain knowledge,are devised to enhance the search efficiency and result quality.Finally,this study validates the effectiveness of Navi through both quantitative experiments and user studies.
