ABSTRACT 
Kappa-P (KnowledgeApplication-Oriented Advanced Database Management 
System) is a parallel database management system running on PIM machines. 
Kappa-P provides efficient database management facilities for knowledge in-
formation processing systems and knowledge base management systems. 

KEY FEATURES 

Nested Relational Model 
  In order to treat complex structured data efficiently, a nested relational 
  model is adopted. With its set constructor and hierarchical attributes, 
  we can represent complex data naturally, and can avoid the unnecessary 
  division of relations. 
Parallel Processing Depending on Data Placement 
  As Kappa-P has a query processing capability at each cluster, the distri-
  bution of relations and the horizontal partition of relations give us inter-
  cluster parallelism. Replication of a relation decentralizes access to the 
  relation. A quasi main memory relation is provided as a replicated rela-
  tion by using intra-cluster parallelism. 
Interaction between Kappa-P and Applications 
  Kappa-P provides functions which reduce communication overhead be-
  tween Kappa-P and various applications. Parts of the applications are 
  executed at clusters at which a query is processed. 

P.75 Figure
Figure: Configuration of Kappa-P
- 75 -