Loading...
Please wait, while we are loading the content...
Similar Documents
Parallelizing datalog: function symbols make a difference (1991).
| Content Provider | CiteSeerX |
|---|---|
| Author | Seib, Jürgen |
| Abstract | ... clause rules. There are, however, some situations de-scribed more easily and more naturally by the use of general terms as arguments. A term is built up from variables, constants, and function symbols. A set of rules where the rules may contain function symbols is called a DatalogFun program. The topic of this paper is parallel processing of decomposable DatalogFUn programs to overcome the performance problem. A decomposable program can be evaluated in parallel such that neither a commu-nication nor a synchronization of the processors has to be established [wS88]. In [SL91] the concept of generalized pivoting is proposed as a sufficient condi-tion for decomposability of arbitrary but function-free Datalog programs. In the current paper the concept of generalized pivoting will be extended to programs which may contain function symbols. |
| File Format | |
| Publisher Date | 1991-01-01 |
| Access Restriction | Open |
| Subject Keyword | Decomposable Datalogfun Program General Term Sufficient Condi-tion Performance Problem Parallel Processing Function-free Datalog Program Current Paper Function Symbol Datalogfun Program Clause Rule Generalized Pivoting Decomposable Program |
| Content Type | Text |
| Resource Type | Article |
