Optimizing the Execution Order of Datalog Rules in VLog

Aus International Center for Computational Logic
Wechseln zu:Navigation, Suche

Optimizing the Execution Order of Datalog Rules in VLog

Vortrag von Alex Ivliev
Datalog extended with existential rules is important for many reasoning tasks. To evaluate a Datalog program, its rules are applied until no new facts can be derived. Crucially, their execution order might significantly impact runtime, memory consumption and even the number of derived facts. Our goal is to find an optimal execution strategy by analysing the relationship between the rules of a given program. We implemented our approach in VLog -- a Datalog rule execution engine -- and measured its impact on performance on several databases.


This talk will take place online via BigBlueButton. To access the room, take one of the following links:

with ZIH-login:

https://selfservice.zih.tu-dresden.de/l/link.php?m=133369&p=679d54f7

without ZIH-login:

https://selfservice.zih.tu-dresden.de/link.php?m=133369&p=e19bcdf1