Search

We consider the problem of real-time log anomaly detection for distributed system with deep neural networks by unsupervised learning. There are two challenges in this problem, including detection accuracy and analysis efficacy. To tackle these two challenges, we propose GLAD, a simple yet effective approach mining for anomalies in distributed systems. To ensure detection accuracy, we exploit the gradient features in a well-calibrated deep neural network and analyze anomalous pattern within log files. To improve the analysis efficacy, we further integrate one-class support vector machine (SVM) into anomalous analysis, which significantly reduces the cost of anomaly decision boundary delineation. This effective integration successfully solves both accuracy and efficacy in real-time log anomaly detection. Also, since anomalous analysis is based upon unsupervised learning, it significantly reduces the extra data labeling cost. We conduct a series of experiments to justify that GLAD has the best comprehensive performance balanced between accuracy and efficiency, which implies the advantage in tackling practical problems. The results also reveal that GLAD enables effective anomaly mining and consistently outperforms state-of-the-art methods on both recall and F1 scores. [ABSTRACT FROM AUTHOR]

The application of novel eco-friendly energy storage ceramics with satisfactory properties is becoming more critical and essential due to environmental threats and energy crises. In this investigation, Na0.5Bi0.5TiO3-based bulk ceramics with an improved recoverable energy storage density and efficiency were prepared by adding a paraelectric compound of SrZr0.5Ti0.5O3 (STZ). As a result, a considerable recoverable energy density (Wrec = 4.32 J cm−3) associated with enhanced energy efficiency (η = 93.48%) was concurrently obtained in the 0.70BBNT–0.30STZ ceramic at 302 kV cm−1. The hidden mechanisms were unveiled through the first-order reversal curve (FORC) and finite element simulations (COMSOL). In addition, the exceptional temperature stability (25–175 °C) of the 0.70BBNT–0.30STZ ceramic was validated at 200 kV cm−1 and a frequency of 20 Hz. More importantly, a transitory discharge time (t0.9 = 92 ns) accompanied by a giant power density (PD = 34.8 MW cm−3) was realized in the 0.70BBNT–0.30STZ ceramic at 120 kV cm−1. These excellent performance features indicate that the 0.70BBNT–0.30STZ ceramic has broad prospects in high pulse power systems. This work proposes a novel strategy for a deeper exploration of energy storage ceramics. [ABSTRACT FROM AUTHOR]

Birefringence in azodye-doped polymethylmethacrylate and azodye-doped silica hybrid materials was induced by the two-photon excitation of a femtosecond laser. The growth and the relaxation characteristics of the photoinduced birefringence in the two types of material were investigated, and the probe transmittance for the induced birefringence was estimated to be 92%. Furthermore, an optical image storage based on this photoinduced birefringence was demonstrated. OCIS codes: 190.0190, 190.4400, 210.0210, 210.4810.

Coumarin 440 doped in sol-gel-derived gel glass was studied for matrix effects and spectral shifts.