GRAND: GAN-based software runtime anomaly detection method using trace information.

Software runtime anomaly detection can detect manifestations (known as anomalies) caused by faults in complex systems before they lead to failure. Whereas most existing methods use external performance indicators, this study uses internal execution traces to reveal failures not only related to software performance issues but also functional errors. A neural network model called GRAND, which combines a variational autoencoder and a generative adversarial network, is proposed to mine anomalies in the execution trace. Cassandra, a widely used database system, was used as a representation to conduct the empirical study. The dataset was collected under a well-designed operational profile that contained 5180 time series, each containing more than ten million data points. GRAND achieved a higher detection performance than the other two SOTA baseline models, with a 99% F1-score compared with 93% and 87%. Ablation studies show that the workload information used in GRAND can determine whether the current internal status is consistent with the task, thus achieving a 16% improvement in the F1-score. The attention mechanism used for data fusion can achieve a 32% improvement in the F1-score.

Screening and identification of multiple abiotic stress responsive candidate genes based on hybrid-sequencing in Vicia sativa.

Common vetch is an important leguminous forage for both livestock fodder and green manure and has a tremendous latent capacity in a sustainable agroecosystem. In the present study, a comprehensive transcriptome analysis of the aboveground leaves and underground roots of common vetch under multiple abiotic stress treatments, including NaCl, drought, cold, and cold drought, was performed using hybrid-sequencing technology, i. e. single-molecule real-time sequencing technology (SMRT) and supplemented by next-generation sequencing (NGS) technology. A total of 485,038 reads of insert (ROIs) with a mean length of 2606 bp and 228,261 full-length nonchimeric (FLNC) reads were generated. After deduplication, 39,709 transcripts were generated. Of these transcripts, we identified 1059 alternative splicing (AS) events, 17,227 simple sequence repeats (SSRs), and 1647 putative transcription factors (TFs). Furthermore, 640 candidates long noncoding RNAs (lncRNAs) and 28,256 complete coding sequences (CDSs) were identified. In gene annotation analyses, a total of 38,826 transcripts (97.78%) were annotated in eight public databases. Finally, seven multiple abiotic stress-responsive candidate genes were obtained through gene expression, annotation information, and protein-protein interaction (PPI) networks. Our research not only enriched the structural information of FL transcripts in common vetch, but also provided useful information for exploring the molecular mechanism of multiple abiotic stress tolerance between aboveground and underground tissues in common vetch and related legumes.

A Dynamic Anomaly Detection Approach Based on Permutation Entropy for Predicting Aging-Related Failures.

Software aging is a phenomenon referring to the performance degradation of a long-running software system. This phenomenon is an accumulative process during execution, which will gradually lead the system from a normal state to a failure-prone state. It is a crucial challenge for system reliability to predict the Aging-Related Failures (ARFs) accurately. In this paper, permutation entropy (PE) is modified to Multidimensional Multi-scale Permutation Entropy (MMPE) as a novel aging indicator to detect performance anomalies, since MMPE is sensitive to dynamic state changes. An experiment is set on the distributed database system Voldemort, and MMPE is calculated based on the collected performance metrics during execution. Finally, based on MMPE, a failure prediction model using the machine learning method to reveal the anomalies is presented, which can predict failures with high accuracy.

The association between short-term response and long-term survival for cervical cancer patients undergoing neoadjuvant chemotherapy: a system review and meta-analysis.

Controversy exists regarding whether a short-term response has an impact on the long-term survival of cervical cancer patients undergoing neoadjuvant chemotherapy (NACT). This study was designed to identify the predictive role of an early response by pooling the results of previous studies. The PubMed and Embase databases were searched through July 2016, and the associations between an early response and disease-free survival (DFS) were pooled by hazard ratio (HR) using random effects models. Six studies involving 490 cervical cancer patients, with 336 responders and 154 non-responders, were finally included in the meta-analysis. The HR for 1-year DFS between early responders and non-responders was 0.25 (95% CI 0.10-0.58, P = 0.001). The HRs for 2-, 3-, 4-, and 5-year DFS were 0.28 (95% CI 0.15-0.56), 0.27 (95% CI 0.16-0.45), 0.29 (95% CI 0.17-0.50) and 0.33 (95% CI 0.20-0.54), respectively. No obvious heterogeneity was found among the studies, with I2 = 0, and a sensitivity analysis showed that all pooled results were robust with logHR confidence limits < 0. An early response was associated with DFS, and responders achieved a significantly higher survival rate than non-responders. This finding should be validated in future prospective studies.