Search

Accurately detecting anomalies in microservice systems is crucial to avoid system failures and economic losses for users. Existing approaches detect anomalies by extracting sequential information from single-modal data, such as metrics or logs. However, they do not clearly analyze the temporal and spatial dependencies of each data type and ignore the correlation between different types of data, which can lead to a significant number of false positives. In this paper, we propose a novel Microservice system Anomaly Detection method via Multi-modal data and Multi-feature extraction (MADMM), which performs a joint analysis of metrics and logs from a spatial and temporal perspective. Specifically, we firstly construct separate feature graphs for metrics and logs in each time window. A graph convolution network is then utilized to capture the spatial correlation among different metrics, while a graph attention network is employed to analyze the contextual relationships among different log events. Then, we design a Cross-Modal Attention-based Gate Recurrent Unit (CMA-GRU) to capture the intricate temporal dependencies of each modal data and facilitate fulfilling cross-modal interactions. Finally, we introduce multi-grained contrastive learning methods to learn robust cross-modal features from both inter- and intra-modality aspects. Experimental results on real-world datasets demonstrate that MADMM outperforms existing baseline methods and exhibits better robustness. [ABSTRACT FROM AUTHOR]

Automated system log anomaly detection plays a crucial role in ensuring service reliability. Existing methods incompletely utilize structured log entries, resulting in the loss of key information such as components and time. Besides, due to the limitations of labeled data, models trained by a single system are difficult to apply to other systems. Therefore, we propose a cross-system log anomaly detection method named LogGT, which simultaneously models log events, components and time, leveraging labeled system knowledge to achieve anomaly detection in unlabeled systems. Specifically, we firstly design a heterogeneous graph to accurately represent the interactions between different events and components in the log sequence. Then, in order to avoid noise interference and conduct cross-system semantic analysis, we employ BERT to extract log sentence vectors, and Normalizing Flow is used to optimize them for smoother node embedding. A Graph Transformer Architecture with Time Intervals (GTAT) is proposed to model heterogeneous graphs by integrating time feature, allowing for a comprehensive analyze of execution order and time anomalies. Additionally, we design a semantic weighting method and utilize a novel domain-adapted transfer learning technology to effectively transfer the heterogeneous graph features of the source system to the target system. Experimental results demonstrate that LogGT outperforms five log anomaly detection methods, achieving an average anomaly detection F1-score higher than 0.95. Moreover, the AUC value of GTAT exceeds the sequence model by more than 2.3%. • A novel log heterogeneous graph modeling method is proposed. • The Normalizing Flow-based BERT model is utilized to generate smooth log semantic. • A Graph Transformer Architecture with Time Interval for order and time anomalies. • A domain-adapted transfer learning with semantic weights for feature transfer. • The comprehensive evaluation of LogGT on three different system datasets. [ABSTRACT FROM AUTHOR]

The maximum shear strain's rotational conditions are investigated under multiaxial loading. The additional hardening is closely related to the relative value, rotational range, and gradual or severe rotation way of the maximum shear strain change. A novel path‐dependent loading factor is introduced to account for the nonproportional additional hardening effect. Subsequently, a computational formula of the structural strain method is established for the multiaxial loading state. The structural stress/strain terms and the proposed path‐dependent factor are introduced into the existing energy‐based fatigue parameter based on the critical plane model. The proposed fatigue prediction method provides prediction results showing that more than 68% and 87% of the data are within the scatter bands of 3 and 5, respectively, for the selected tube–flange/tube welded joints under multiaxial loading paths. Highlights: Rotational conditions of maximum shear strain affect nonproportional hardening degree.Path factor should incorporate effect of gradual/severe rotation of maximum shear strain.Multiaxial structural strain method is established by equilibrium condition and flow rule.Energy‐based FPw including path factor provides satisfactory prediction for welded joints. [ABSTRACT FROM AUTHOR]

• Low-cycle multiaxial fatigue experiments are conducted for Q235B steel welded material. • Equivalent cyclic stress-strain curves and hysteresis loops are obtained. • Specimens under non-proportional paths show certain extent of additional hardening effect. • Proposed fatigue parameter incorporates energy on weighted critical plane and orthogonal plane. • The proposed fatigue parameter provides a superior ability to make life prediction. Low-cycle torsional and multiaxial fatigue experiments are carried out for thin-walled Q235B steel welded specimens under both in-phase and 90° out-of-phase loading paths. A remarkable non-proportional additional hardening effect exists in the Q235B steel welded material under the 90° out-of-phase loading paths. A new energy-based fatigue parameter is proposed that incorporates the shear strain energy on the weighted critical plane and the normal strain energy on both the weighted critical plane and orthogonal plane. The proposed energy-based fatigue parameter provides a superior ability to make life prediction for the Q235B steel welded material under the torsional and multiaxial loading paths. [ABSTRACT FROM AUTHOR]

Existing Web services recommendation models suffer from matrix sparsity and cold-start problems due to the limited number of user-service interactions and too little contextual information, so this paper introduces knowledge graph as auxiliary information to Web service recommendation to alleviate matrix sparsity and cold-start problems. And we propose the Web services recommendation based on Metapath-guided Graph Attention Network Model (WSR-MGAT) to fully exploit the structural information of the knowledge graph to improve the recommendation accuracy. Specifically, WSR-MGAT uses graph embedding method to obtain the initial embedding of entities and relationships. In order to obtain more closely related neighbors, we propose to use a distance-aware path sampling method to extract meta-path instances with closer relationships. Previous knowledge graph-based Web service recommendations do not make full use of the rich interaction information, which may lead to limited performance. To address this problem, this paper uses meta-paths to guide nodes to recursively aggregate higher-order neighbor information and use an attention mechanism to distinguish the importance of neighbors. Meanwhile, the semantic information of nodes under different meta-paths is fused to obtain a more comprehensive embedding of nodes. To verify the performance of the model, we use the real data crawled from the ProgrammableWeb platform to conduct multiple groups of experiments. The experimental results show that the WSR-MGAT improves over the strongest baselines w.r.t. Prec@10 by 4.6%; Recall@10 by 3.8%; NDCG@10 by 4.2% and F1@10 by 4.9%. [ABSTRACT FROM AUTHOR]

Corrosion fatigue (CF) experiments of G20Mn5QT cast steel and Q355D steel butt-welded joint are conducted with loading frequencies of 1 Hz and 10 Hz. The decrease of loading frequency resulting in a reduction in the CF life of specimens. The average percentage of CF life reduction are 6.4% (S = 300 MPa), 53.5% (S = 260 MPa), 17.5% (S = 220 MPa), 34.3% (S = 190 MPa), 38.6% (S = 160 MPa) and 80.5% (S = 130 MPa), respectively. The CF failure mechanism remains unchanged regardless of the loading frequency. Based on continuum damage mechanics (CDM), a novel CF life prediction model is proposed, in which the effect of loading frequencies is introduced by the evolution of corrosion pits. The prediction model considers the coupling effect of fatigue and corrosion, and it achieves the prediction of CF life of the butt-welded joints using experiment data at higher loading frequency. The accuracy and universality of the prediction model are verified by predicting the CF life of other specimens. [Display omitted] • The S–N curves of butt-welded joints under different loading frequencies is obtained. • The corrosion fatigue failure mechanism of butt-welded joints is analyzed. • A corrosion fatigue life prediction model considering loading frequency is proposed. • The prediction of corrosion fatigue life of the butt-welded joints is achieved. • The accuracy and universality of the prediction model are verified. [ABSTRACT FROM AUTHOR]

Most existing Web service recommendation models based on machine learning do not fully consider the high-order features interaction between users and services and with poor interpretability. In this paper, an Interpretable Web Service Recommendation model based on Disentangled Representation Learning (WSR-DRL) is proposed. First of all, to make full use of the service description information to improve the accuracy of Web service recommendation, the features representation of service name is obtained by using BERT model, and the local and global features representation of service description information is further obtained by combining 2-D CNN and Bi-LSTM. Then the disentangled convolution neural network is used to generate the high-order interaction features between users and services, and the neighborhood routing algorithm is used to mine the latent factors in these features. That improves the accuracy of Web service recommendation and make it interpretable. Finally, in order to verify the effectiveness of the model, several groups of experiments are carried out on real data sets. The experimental results show that compared with latest models such as DMF, DeepFM, DKN, GCMC, NDCG model and WSR-MGAT model, the WSR-DRL model proposed in this paper shows better performance on Precision@10, Recall@10, F1@10 and NDCG@10 evaluation metrics. [ABSTRACT FROM AUTHOR]

System logs are an important data source for performance monitoring and anomaly detection. Analyzing logs for anomaly detection can improve service quality. At present, although machine learning algorithms for anomaly detection can achieve high accuracy, they lack robustness. The detection model cannot dynamically adapt to changes of logs when system logs contain noises owing to the casualness of the operators or log templates update. In face of this challenge, the paper proposes a robust log anomaly detection method based on contrastive learning and multi-scale Masked Sequence to Sequence (MASS). First, a log feature extraction model integrating the BERT model with contrastive learning is proposed. It can extract effective features by pulling two related normal logs together and pushing apart normal and abnormal logs to ensure that the semantic similarity between normal and abnormal log templates is lower than that between normal log templates, effectively remove abnormal log templates and distinguish log categories to which normal log templates and normal noise log templates belong rather than rudely treating the noise log templates as anomalies, which enhances the robustness of anomaly detection. Then, a multi-scale Masked Sequence to Sequence (MSMASS) model is proposed, the Attention mechanism of the MASS model is replaced with multi-scale Attention to fully learn the context information of different scales of the log sequence, which improves the accuracy of anomaly detection. Contrast experiments are conducted with four baseline methods on common datasets, and the results show that the method proposed in this paper is superior to most existing log-based anomaly detection methods in terms of accuracy and robustness. [ABSTRACT FROM AUTHOR]