Survivability of Cloud Databases – Factors and Prediction

  • Jose Picado
  • Willis Lang
  • Edward C. Thayer

International Conference on Management of Data |

Published by ACM

Download PDF | View Publication | View Publication | View Publication

Public cloud database providers observe all sorts of different usage patterns and behaviors while operating their services. Service providers such as Microsoft try to understand and characterize these behaviors in order to improve the quality of their service, provide new features for customers, and/or increase the efficiency of the operations. While there are many types of patterns of behavior that are of interest to providers, such as query types, workload intensity, and temporal activity, in this paper, we focus on the lowest level of behavior — how long do public cloud databases survive before being dropped? Given the large and diverse relational database population that Azure SQL DB has, we present a large-scale survivability study of our service and identify some factors that can demonstrably help predict the lifespan of cloud databases. The results of this study are being used to influence how Azure SQL DB operates in order to increase efficiency as well as improve customer experience.