Conclusion - Information Systems Reengineering, Integration and Normalization - page 373

all of the original conceptual schema semantics. With user input, we can provide a
relational schema that is closer to the user expectation and preserves the existing
schema constraints such as record key, records relationships, and attributes.
For data conversion, the methodology provides algorithms to unload a hierarchi-
cal or network database into sequential files directly and effectively, with minimum
user involvement. These files can then be uploaded onto the target system with little
additional effort.
In program translation, the methodology provides an “open” data structure by
adding secondary indices to the existing hierarchical or network database. This
eliminates the navigation access path required to retrieve a target record from a sys-
tem record. Instead, each target record type can be accessed directly without data-
base navigation. The database access time is thus reduced and the program conver-
sion effort simplified. The methodology also provides algorithms to translate SQL
statements into hierarchical or network DML statements. These are sound solutions
to the program conversion problem.
Basically, the methodology is similar to the relational interface approach in that
both provide a relational interface to make the hierarchical or network DBMS a
relational-like DBMS. The methodology can help the users in the following ways:
• Applythemethodologytoconvertahierarchicalornetworkdatabasesystem
into relational database system.
The methodology is an integrated approach to solve the conversion
problem. The user has a solution for the whole task.
• Applypartofthemethodologytoreduceconversionproblems.
The methodology includes schema translation, data conversion, and program
translation. Each process can be applied
independently as required.
• Applyschematranslationtoconstructadistributeddatabasesystem.
In a distributed database system, many local schema act
independently for their own local applications. To implement a
major application or a global application, we must integrate this schema into a
global schema. Our methodology is used to obtain a common EER model for a
number of local hierarchical or network
schema.
• Applythetechniqueofaddingsecondaryindicestoprovidean“open”structure
database gateway.
Currently, many vendors provide database gateways to allow other
vendors' database programs to access their databases. The
addition of secondary indices is an alternative approach.
• Applythemethodologyforamoreuser-friendlyinterfacetoend-users.
The methodology is used to provide a relational interface to a
nonrelational system. It allows a company to continue using a
network or hierarchical while, at the same time, users can use
the friendly interface supported by a relational database.