Distributed Database Architecture
Data independence:
Logical data independence:
User application should not be affected by the logical data independence.
Physical independence:
Hiding the details of storage structure.
only the centralized DBMS form is important.
Naming Transparency:
location transparency:
An operation performed on the data is independent of both location and system where it is executed.
Naming transparency :
Unique name is assigned to each identifiers.
Unique name is assigned to each identifiers.
Transparency:
Higher level semantics is separated from the lower level semantics.
Hiding implementation from the users.
Forms:
Network transparency
data transparency
fragmentation transparency
replication transparency
Transparency can be provided at different level of the systems.
Replication transparency:
Data is replicated for reliability and performance consideration.
Fragment transparency:
DB relation are divided into smaller fragments for performance and availability reasons.
The origin of transparency:
Access layer : Transparency feature are built into user language that is translated into requested operations
DBMS: act as an integrated OS and DBMS,make necessary translation between OS and user interface
OS: transparent access is provided to resources.
DBMS architecture:
Based on components: the components and their interrelations are defined.
Based on functions: The functions that the system will perform are defined.
Based on data: Different type of data and functional unit that will use data are defined/
Usually there is an interplay among them.
Architecture:
Architecture of a system defines the structure.
Model:
Model:
In the foreign part of the structure there is the reference model.
it is idealized view,
it is different from real world example but gives reasonable framework.
Goal:
present the issues that need to be addressed.
present a framework with in which the design and implementation can be discussed.
ANSI model:
A model from the 70's but still provide the interest and is the basis of distributed database systems.
Bases on data organization:
External view: highest level of user view of the portion and relationship among data.
conceptual view: Abstract definition of the database the real world view.
internal view: lowest level physical definition storage issue.
Comments
Post a Comment