An object flow computer for database applications: Design and performance evaluation
This paper presents the design and evaluation of an Object Flow Computer (OFC) for efficiently support the processing of object-oriented databases. Object Flow Processing is presented as a special case of data flow processing in that objects are the primary data flow through the system. A general OFC architecture is shown; in the OFC, the objects originate from base classes and flows to processing nodes. Architecture of processing nodes is also described; each node consists of a general-purpose processor, zero or more special-purpose coprocessors, memory and secondary storage devices. The set of system defined OFC operators used in the object identification phase contains: Selector, Sorter, Join (all with some variants), Constructors, Sets Operators. The query processing in OFC consists of two phases: identifying the qualified objects in the database and then sending messages to the identified objects to invoke the execution of the corresponding methods. In the section dedicated to the performance evaluation the author shows that the two-phase parallel query processing strategy is more efficient than the conventional one, in processing complex query.
- Database machines. 6th international workshop, IWDM '89, Deauville, France, June 19-21, 1989. Proceedings
- Parallel execution of object functional queries
- An Experimental Performance Evaluation of Join Algorithms for Parallel Object Databases
- Performance evaluation of a parallel cascade semijoin algorithm for computing path expressions in object database systems
- New concepts for parallel object-relational query processing
This page was built for publication: An object flow computer for database applications: Design and performance evaluation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2365577)