Accurately identifying time-invariant operational relationships among different components is critical to autonomic management of complex manufactural systems. In this paper, we collect time series of sensor readings from manufacturing systems, and propose a solution leveraging Sparse Group LASSO to discover structured pairwise nonlinear relationships and quantify them by mathematical formulas. We consider both real-life operational patterns and underlying physical reactions inside the manufactural systems, which leads to a learning formulation for combined periodic and aperiodic system behaviors. An accelerated gradient descent algorithm is developed to efficiently solve the related optimization problem. We estimate sample correlations between proximal time points to improve the accuracy of the discovered relationships and the nonlinear quantitative formulas. The method is evaluated using both synthetic and real- world datasets, which shows superior performance over the state of the art in discovering nonlinear relationships in manufactural systems.

Sparse coding plays a key role in high dimensional data analysis. One critical challenge of sparse coding is to design a dictionary that is both adaptive to the training data and generalizable to unseen data of same type. In this paper, we propose a novel dictionary learning method to build an adaptive dictionary regularized by an a-priori over-completed dictionary. This leads to a sparse structure of the learned dictionary over the a-priori dictionary, and a sparse structure of the data over the learned dictionary. We apply the hierarchical sparse dictionary learning approach on both synthetic data and real-world high-dimensional time series data. The experimental results demonstrate that the hierarchical sparse dictionary learning approach reduces overfitting and enhances the generalizability of the learned dictionary. Moreover, the learned dictionary is optimized to adapt to the given data and result in a more compact dictionary and a more robust sparse representation. The experimental results on real datasets demonstrate that the proposed approach can successfully characterize the heterogeneity of the given data, and leads to a better and more robust dictionary.

An accurate and automated identification of operational behavior switching is critical to the autonomic management of complex systems. In this paper, we collect sensor readings from those systems, which are treated as time series, and propose a solution to discover switching behaviors by inferring the relationship changes among massive time series. The method first learns a sequence of local relationship models that can best fit the time series data, and then combines the changes of local relationships to identify the system level behavior switching. In the local relationship modeling, we formulate the underlying switching identification as a segmentation problem, and propose a sophisticated optimization algorithm to accurately discover different segments in time series. In addition, we develop a hierarchical optimization strategy to further improve the efficiency of segmentation. To unveil the system level behavior switching, we present a density estimation and mode search algorithm to effectively aggregate the segmented local relationships so that the global switch points can be captured. Our method has been evaluated on both synthetic data and datasets from real systems. Experimental results demonstrate that it can successfully discover behavior switching in different systems.

This paper proposes a novel framework to automatically pinpoint suspicious sensors that lead to the quality change in physical systems such as manufacture plants. Our framework treats sensor readings as time series, and contains three main stages: time series transformation to feature series, feature ranking, and ranking score fusion. In the first step, we transform time series into a number of different feature series to describe the underlying dynamics of each sensor data. After that, the importance scores of all feature series are computed by utilizing several feature selection and ranking techniques, each of which discovers specific aspects of feature importance and their dependencies in the feature space. Finally we combine importance scores from all the rankers and all the features to obtain the final ranking of each sensor with respect to the system quality change. Our experiments based on synthetic time series as well as sensor data from a real system demonstrate the effectiveness of proposed method. In addition, we have implemented our framework as a production engine, and successfully applied it to several real physical systems.

Performance bugs which manifest in a production cloud computing infrastructure are notoriously difficult to diagnose because of both the difficulty of reproducing those bugs and the lack of debugging information. In this paper, we present PerfScope, a practical online performance bug inference tool to help the developer understand how a performance bug happened during the production run. PerfScope achieves online bug inference to obviate the need for offline bug reproduction. PerfScope does not require application source code or any runtime instrumentation to the production system. PerfScope is application-agnostic, which can support both interpreted and compiled programs running inside a cloud infrastructure. We have implemented PerfScope and tested it using real performance bugs on seven popular open source server systems (Hadoop, HDFS, Cassandra, Tomcat, Apache, Lighttpd, MySQL). The results show that PerfScope can narrow down the search scope of the bug-related functions to a small percentage (0.03-2.3%) and rank the real bug-related functions within top five candidates in the majority of cases. PerfScope only imposes on average 1.8% runtime overhead to the tested server applications.

Logs are ubiquitous for system monitoring and debugging. However, there lacks a comprehensive system that is capable of performing heterogeneous log organization and analysis for various purposes with very limited domain knowledge and human surveillance. In this manuscript, a novel system for heterogeneous log analysis is proposed. The system, denoted as Heterogeneous Log Analyzer (HLAer), achieves the following goals concurrently: 1) heterogeneous log categorization and organization; 2) automatic log format recognition and 3) heterogeneous log indexing. Meanwhile, HLAer supports queries and outlier detection on heterogeneous logs. HLAer provides a framework which is purely dataoriented and thus general enough to adapt to arbitrary log formats, applications or systems. The current implementation of HLAer is scalable to Big Data.