ARIES (computer science)

from Wikipedia, the free encyclopedia

ARIES ( Algorithms for Recovery and Isolation Exploiting Semantics ) is a family of algorithms for recovery of database systems after a failure.

After an error, for example due to a crash or a hardware failure, the content permanently stored in a database may be inconsistent (example: money has already been debited from account X, but not yet posted to account Y.). These sources of error must therefore be excluded; ARIES contains possible solutions.

ARIES is based on a so-called no-force / steal strategy, i.e. H. Changed database pages are not necessarily written to persistent storage media at the end of a transaction ( no-force ) and database pages that are still being used by an ongoing transaction can still be swapped out prematurely ( steal ). ARIES also uses the so-called "Write Ahead Logging" ( WAL principle ), i. H. Modifications are logged before the actual writing.

literature

  • C. Mohan: Repeating History Beyond ARIES. ( Memento of March 18, 2009 in the Internet Archive ) (PDF; 282 kB). In: Malcolm P. Atkinson, Maria E. Orlowska, Patrick Valduriez, Stanley B. Zdonik, Michael L. Brodie (Eds.): Proceedings of the 25th International Conference on Very Large Data Bases. Edinburgh, Scotland, UK, September 7th - 10th 1999. Morgan Kaufmann Publishers Inc., Orlando FL 1999, ISBN 1-55860-615-7 , pp. 1-17.

Web links