Font Size: a A A

Relational Database Management System, Data Dictionary And Implementation

Posted on:2005-09-08Degree:MasterType:Thesis
Country:ChinaCandidate:S F ZhangFull Text:PDF
GTID:2208360122470011Subject:Computer software and theory
Abstract/Summary:PDF Full Text Request
Database software is the kernel of data processing,our scientists are always trying to develop our own database management system,and the same time the vital department such as national defense,govement,finace demand an independent and safe DBMS.So developing our own DBMS will effect Chinese software industry and protect national information.This paper is about the research work for "DBMS AND APPLICATION" project which is a national 863 high tech program, introducing the struct of catalog and cache.lt also desiged and implemented all functions of catalog, supporting the system working in good order and with high performance..Catalog is one of the most important of DBMS,which is a read-only set of tables that provides information about the database.Creating the system tables and importing the basic system records when creating a new database by a simple parser, initializing the cache struct and loading system records befor opening the database for normal operation.It must keep the data constency when define, modify,drop a schema object, and save the dependecy in catalog, the dependency will be used when drop or modify a shema object The catalog is structured in tables and viewsjust like other database data. Not only is the catalog central to every database, it is an important tool for all users, from end users to application designers and database administrators. Use SELECT SQL statements to access the catalog. Data in the base tables of the catalog is necessary for dbms to function. Much of the catalog information is kept in the catalog cache,because DBMS constantly accesses the catalog during database operation to validate user access and to verify the state of schema objects. All information is stored in cache using the hash algorithm.Concurrency control, cache consistancy, replace algorithm is the central of share cache. Introducing a changed lock based concurrency control model to adapt to the muti-user and muti-cpus enverment, a new cache consistency for DBMS catalog. This paper thoroughly discusses the design and implementation of the catalog for OSCAR DBMS ,getting a good result in performance and scalability.
Keywords/Search Tags:catalog, cache, dbms, concurrency, consistency, replace
PDF Full Text Request
Related items