With a new era of cloud and big data, Database Management Systems (DBMSs)have become more crucial in numerous enterprise business applications in allthe industries. Accordingly, the importance of their proactive and preventivemaintenance has also increased. However, detecting problems by predefined rulesor stochastic modeling has limitations, particularly when analyzing the data onhigh-dimensional Key Performance Indicators (KPIs) from a DBMS. In recentyears, Deep Learning (DL) has opened new opportunities for this complexanalysis. In this paper, we present two complementary DL approaches to detectanomalies in SAP HANA. A temporal learning approach is used to detect abnormalpatterns based on unlabeled historical data, whereas a spatial learningapproach is used to classify known anomalies based on labeled data. Weimplement a system in SAP HANA integrated with Google TensorFlow. Theexperimental results with real-world data confirm the effectiveness of thesystem and models.

Deep Learning on Key Performance Indicators for Predictive Maintenance in SAP HANA

Scan to connect with one of our mobile apps

Coinbase Wallet app

Coinbase app

Or try the Coinbase Wallet browser extension