Abstract
This paper presents the algorithms necessary to partition an Extended Predicate connection Graph (EPCG) that represents a deductive database for use with a distributed system. The algorithms satisfy our initial partitioning goals which include equalization of storage and processing costs, distributing the base relation nodes in an effective manner and preserving the locality of nodes within the EPCG as much as possible. These algorithms also account for multiple users who may be simultaneously accessing the system from various 'share nothing' workstations that are connected via a communication network.
Original language | English (US) |
---|---|
Title of host publication | Proceedings - ACM Computer Science Conference |
Editors | Dawn Cizmar |
Publisher | Publ by ACM |
Pages | 288-295 |
Number of pages | 8 |
ISBN (Print) | 0897916344 |
State | Published - Jan 1 1994 |
Event | Proceedings of the 22nd Annual ACM Computer Science Conference - Phoenix, AZ, USA Duration: Mar 8 1994 → Mar 10 1994 |
Other
Other | Proceedings of the 22nd Annual ACM Computer Science Conference |
---|---|
City | Phoenix, AZ, USA |
Period | 3/8/94 → 3/10/94 |
ASJC Scopus subject areas
- Hardware and Architecture