The growing size of today’s data centers and the expectation of 24/7 availability continuously increase the complexity of hardware administration. To this end, the Self-Monitoring, Analysis, and Reporting Technology has been developed to provide insights into the health state of hard disk drives. Many approaches to predicting hard disk drive failures based on such monitoring data have been proposed in recent years. Nevertheless, most approaches consider this problem only as a static task, i.e., they train a static machine learning model on a given training set and evaluate its performance on a test set. However, due to model aging and changes in failure patterns, previously learned prediction models must be updated during runtime, which requires a time-dependent evaluation. Therefore, we present four machine learning model updating strategies, build multiple models for hard disk drive failure prediction using four machine learning algorithms, and compare the prediction quality of the different model update strategies and machine learning algorithms. Experimental results using a real-world data set of hard disk drives demonstrate the need for model update strategies, with XGBoost using the Hoeffding bound update trigger achieving the overall best prediction performance concerning prediction quality and number of updates required.

Predictive maintenance is an essential aspect of the concept of Industry 4.0. In contrast to previous maintenance strategies, which plan repairs based on periodic schedules or threshold values, predictive maintenance is normally based on estimating the time-to-failure of machines. Thus, predictive maintenance enables a more efficient and effective maintenance approach. Although much research has already been done on time-to-failure prediction, most existing works provide only specialized approaches for specific machines. In most cases, these are either rotary machines (i.e., bearings) or lithium-ion batteries. To bridge the gap to a more general time-to-failure prediction, we propose a generic end-to-end predictive maintenance methodology for the time-to-failure prediction of industrial machines. Our methodology exhibits a number of novel aspects including a universally applicable method for feature extraction based on different types of sensor data, well-known feature transformation and selection techniques, adjustable target class assignment based on fault records with three different labeling strategies, and the training of multiple state-of-the-art machine learning classification models including hyperparameter optimization. We evaluated our time-to-failure prediction methodology in a real-world case study consisting of monitoring data gathered over several years from a large industrial press. The results demonstrated the effectiveness of the proposed methodology for six different time-to-failure pre-diction windows, as well as for the downscaled binary prediction of impending failures. In this case study, the multi-class feed-forward neural network model achieved the overall best results.

In many areas of decision making, forecasting is an essential pillar. Consequently, there are many different forecasting methods. According to the "No-Free-Lunch Theorem", there is no single forecasting method that performs best for all time series. In other words, each method has its advantages and disadvantages depending on the specific use case. Therefore, the choice of the forecasting method remains a mandatory expert task. However, expert knowledge cannot be fully automated. To establish a level playing field for evaluating the performance of time series forecasting methods in a broad setting, we propose Libra, a forecasting benchmark that automatically evaluates and ranks forecasting methods based on their performance in a diverse set of evaluation scenarios. The benchmark comprises four different use cases, each covering 100 heterogeneous time series taken from different domains. The data set was assembled from publicly available time series and was designed to exhibit much higher diversity than existing forecasting competitions. Based on this benchmark, we perform a comprehensive evaluation to compare different existing time series forecasting methods.

The complexity of managing the capacities of large IT infrastructures is constantly increasing as more network devices are connected.This task can no longer be performed manually, so the system must be monitored at runtime and estimations of future conditions must be made automatically.However, since using a single forecasting method typically performs poorly, this paper presents a framework for forecasting univariate network device workload traces using multiple forecasting methods.First, the time series are preprocessed by imputing missing data and removing anomalies.Then, different features are derived from the univariate time series, depending on the type of forecasting method.In addition, a recommendation approach for selecting the most suitable forecasting method from this set of algorithms for each time series based only on its historical values is proposed.For this purpose, the performance of the forecasting methods is approximated using the historical data of the respective time series under consideration.The framework is used in the FedCSIS 2020 Challenge and shows good forecasting quality with an average R 2 score of 0.2575 on the small test data set.

One central problem of machine learning is the inherent limitation to predict only what has been learned -stationarity. Any time series property that eludes stationarity poses a challenge for the proper model building. Furthermore, existing forecasting methods lack reliable forecast accuracy and time-to-result if not applied in their sweet spot. In this paper, we propose a fully automated machine learning-based forecasting approach. Our Telescope approach extracts and transforms features from an input time series and uses them to generate an optimized forecast model. In a broad competition including the latest hybrid forecasters, established statistical, and machine learning-based methods, our Telescope approach shows the best forecast accuracy coupled with a lower and reliable time-to-result.

Due to the fast-paced and changing demands of their users, computing systems require autonomic resource management. To enable proactive and accurate decision-making for changes causing a particular overhead, reliable forecasts are needed. In fact, choosing the best performing forecasting method for a given time series scenario is a crucial task. Taking the "No-Free-Lunch Theorem" into account, there exists no forecasting method that performs best on all types of time series. To this end, we propose an automated approach that (i) extracts characteristics from a given time series, (ii) selects the best-suited machine learning method based on recommendation, and finally, (iii) performs the forecast. Our approach offers the benefit of not relying on a single method with its possibly inaccurate forecasts. In an extensive evaluation, our approach achieves the best forecasting accuracy.

Modern distributed systems and Internet-of-Things applications are governed by fast living and changing requirements. Moreover, they have to struggle with huge amounts of data that they create or have to process. To improve the self-awareness of such systems and enable proactive and autonomous decisions, reliable time series forecasting methods are required. However, selecting a suitable forecasting method for a given scenario is a challenging task. According to the “No-Free-Lunch Theorem,” there is no general forecasting method that always performs best. Thus, manual feature engineering remains to be a mandatory expert task to avoid trial and error. Furthermore, determining the expected time-to-result of existing forecasting methods is a challenge. In this article, we extensively assess the state-of-the-art in time series forecasting. We compare existing methods and discuss the issues that have to be addressed to enable their use in a self-aware computing context. To address these issues, we present a step-by-step approach to fully automate the feature engineering and forecasting process. Then, following the principles from benchmarking, we establish a level-playing field for evaluating the accuracy and time-to-result of automated forecasting methods for a broad set of application scenarios. We provide results of a benchmarking competition to guide in selecting and appropriately using existing forecasting methods for a given self-aware computing context. Finally, we present a case study in the area of self-aware data-center resource management to exemplify the benefits of fully automated learning and reasoning processes on time series data.