Browsing by Author "Landim, Lauriana Patricia Tavares"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
- Monitoring and detection of anomaly in microservices environmentsPublication . Landim, Lauriana Patricia Tavares; Lopes, Eurico Ribeiro; Barata, Luís Miguel Santos Silva AscensãoMicroservices architectures have become increasingly popular in recent years because of their scalability and agility. However, the distributed nature of this architecture also introduces some challenges, especially in terms of monitoring and detecting anomalies. Anomaly detection is the process of identifying anomalous events or patterns in data that do not conform to expected behavior. In microservices environments, this eventually becomes very important, since the number of services tends to grow increasingly, making the interaction between them complex. Because it is recent, there are still few studies on the best approaches to detecting anomalies in microservices. This thesis investigates how well PyOD library algorithms can detect anomalous behavior in a microservices dataset. PyOD is an open-source Python toolbox for performing scalable outlier detection on multivariate data. Some benefits of PyOD are that it is scalable, includes several algorithms, and can detect anomalies in multivariate data. We also review among the PyOD, KNN and HBOS algorithms, which one performs better at detecting anomalies. To evaluate the approach, we used TraceRCA dataset to detect anomalies such as application bugs, CPU exhausted, and network jam. This dataset contains logs from a real microservices system. The preliminary results show that the HBOS algorithm performs better than kNN, with Recall and F1-Score of 83% and 91%, respectively, while for kNN these metrics were 80% and 89%, respectively.
- A simple approach to detect anomalies in microservices-based systems using PyODPublication . Landim, Lauriana Patricia Tavares; Barata, Luís; Lopes, EuricoEase of scale is one of the defining characteristics of microservices. However, with scalability comes the problem of diversity of services, making it very important to detect anomalies the soonest possible. Because it is recent, there are still few studies on the best approaches to detecting anomalies in microservices. This paper proposes the Python toolkit, PyOD, as an approach for microservice anomaly detection. This toolkit is composed of a set of anomaly detection algorithms, including classical LOF (SIGMOD2000) to the latest ECOD (TKDE2022). To evaluate the approach, we used two of its algorithms, k Nearest Neighbors (kNN) and Histogram-based Outlier Score (HBOS) to detect anomalies such as application bugs, CPU exhausted, and network jam on the TraceRCA dataset. This dataset contains logs from a real microservices system. The preliminary results show that HBOS algorithm performs better than kNN, with Recall and F1-Score of 93% and 89%, respectively, while for kNN these metrics were 92% and 85%, respectively.