Processing Distributed Set Queries : The Set Equality Correlated Case

スポンサーリンク

概要

• 論文の詳細を見る

• To build an efficient distributed system, problems associated with distributed databases have to be solved. Among these problems, the problem of efficient query processing is important. In this paper, we discuss efficient processing of set queries. Since processing set queries involves transmition of large volumes of data, finding efficient ways to process these queries is important. We discuss set queries in context of SQL, the approach, however, is general. A set query in SQL is represented by the nested form ; i.e. as two query blocks connected by a set operation. The objective is to minimize transmission cost by minimizing size of data to be transmitted between sites in the system. The approach is similar to the one in case of centeralized set queries. Functional dependencies between attributes in the query are used to minimize size of transmitted data. A set query is transformed into either a distributed non-set query or a local set query. Optimization of distributed nested SQL queries was discussed in [2], however, optimization of set SQL queries is missing. Optimization of set queries, in general, was discussed in [3]. The objective is to find a transmission plan that minimizes the communication cost. It is assumed that relations in the database are partitioned. Set queries are classified into two main types, correlated and uncorrelated. A query is correlated when the inner query block refers to a relation in the outer block, and uncorrelated otherwise. Here, correlated set queries are considered. Correlated queries are then classified into two types, one when the WHERE clause of the outer block contains a constant, the second when it contains an attribute. These types are called type-SCOR1 and type-SCOR2, respectively. The general form of these two types are : Type-SCORE1 SERECT RLC FROM RI WHERE "Constant" SET OPERATION SELECT RJ. K FROM RJ WHERE RI. V = RJ. U Type-SCOR2 SELECT RI. C FROM RI WHERE RI. K SET OPERATION SELECT RJ. K FROM RJ WHERE RI. V = RJ. U. Here, we concentrate on queries where the set operation is =ALL(i.e. set equality operation).

• 一般社団法人情報処理学会の論文
• 1989-03-15

著者

• Kambayashi Yahiko

  Department of Social Informatics, Kyoto University

• Kambayashi Yahiko

  Department Of Social Informatics Kyoto University

• Kambayashi Y

  Kyoto Univ. Kyoto‐shi Jpn

• El-Sharkawi Mohamed

  九大

• Kambayashi Yahiko

  九大

• El‐sharkawi M

  Kyushu Univ.

関連論文

• 内容および構造に基づくXML文書のアクセス権の変換について(セッション1B : XML基礎技術)
• 内容および構造に基づくXML文書のアクセス権の変換について(XML基礎技術)(「夏のデータベースワークショップ(DBWS2003)」一般)
• Object Migration in Object-Oriented Databases
• Translating Access Authorizations for Transformed XML Documents
• Subset Graphs and Their Applications to Databases (Applied Combinatorial Theory and Algorithms)
• Cooperative Query Formulation for Geographic Databases
• Processing Distributed Set Queries : The Set Equality Correlated Case

スポンサーリンク