In many application areas it is useful to not only keep and evaluate the current state of data, but also its history, thus enabling temporal data management and temporal analytics, but also requiring scalable database management systems to cope with the data volumes and performance requirements.

The inclusion of temporal operations in the SQL:2011 standard as well as the ever-decreasing price of memory (RAM) as well as storage (disk/flash) have renewed interest in temporal databases. Existing solutions, however, are rather limited: Commercial systems as well as approaches from the research community support only a small set of temporal operations with rather limited performance, not incorporating the advances in modern database architecture such as main-memory databases.

In collaboration with partners from the research community and industry we are developing data structures and operators that are designed for massively parallel main-memory databases, support a wide range of operators and provide very high and predictable performance.