Abstract
Database schema evolution is one of the grand challenges in data management research and practice. In this paper, we survey available schema evolution benchmarks. We argue that existing benchmarks do not reflect the more recent demands on database applications, such as online data migration for high availability applications, agile release strategies, and schema changes that affect more than one ...
Abstract
Database schema evolution is one of the grand challenges in data management research and practice. In this paper, we survey available schema evolution benchmarks. We argue that existing benchmarks do not reflect the more recent demands on database applications, such as online data migration for high availability applications, agile release strategies, and schema changes that affect more than one data model. We conclude that a new generation of schema evolution benchmarks is called for, which reflect these new demands, especially in the context of schema evolution in NoSQL data stores.