fbpx

DATABASE INTEGRATION TECHNIQUES

If different multiple applications have stored their data in specific database called integration  database ,it is called database integration.Because of that data can be available across all different applications.Data can be accessed easily and there is no need to transfer across different applications.

There are several database integration topics,among them we are going to discuss some topics.

Integration by Batch copy-Data is extracted from external sources as the initial step and it is loaded into GIS.This approach differs according to sophistication of its implementation.first,the data is manually extracted then transferred by tape or other media and manually loaded in less sophisticated implementations.But in the more sophisticated implementations,it is the basis for data warehousing.There,data is automatically extracted,transferred and loaded on  a regular and frequent basis.

Advantages-cost:special integration products are not required often.database management system tools  for importing and exporting data can be used.It requires only minimal database changes.

Disadvantages-data currency .Even though batch copy is fully automated.It is done at longer basis or overnight.When we need take expensive business decisions according to stale data.This issue becomes more critical.

Data access,data integrity and scalability-GIS system is the only thing which has access to  the shared data.Other systems haven’t.As,GIS is working with snapshot copy-data access must have to read-only for users and applications to ensure data integrity.

Application Level Integration

It performs the integration at application level.Application can be modified to retrieve information directly from other systems when the applications are  able to communicate with multiple databases directly.

This stands good against data currency and administration.because of the direct access of data in their native databases ,the data will be always current.The changes to the database are minimal.

This technique has poor encapsulation.Any data synthesis has to be implemented  within application because the integration is in application level.If the integrated application is going to update multiple databases there we have to manage transactions that apply across databases.This technique is good for few databases.If its more it will fail.The cost for the initial implementation is low and overtime it increases due to encapsulation problems.Data access is average.

Application level integration with middleware

It also do application level database application.Difference between this and above mentioned technique is addition of middleware.Some application may be unable to communicate directly

depending on the Database Management software they are using.But,middleware offers mechanism for the application to communicate with the database which does  not supported by application.

As now  application is having one more layer than directly working with data,there is performance penalty.

Cost is increased due to purchase of middleware products.some middleware can perform data synthesis.Some can provide interface to multiple databases.

Database level integration

It implements at the database level instead of application level.External data is presented by GIS to GIS applications  as if the data is stored locally in the GIS.They may directly communicate or use middleware depending on databases.

Advantages-Encapsulation is much better because data location is transparent to the applications and applications can access data as it were stored in GIS database.Advantages are in scalability,data integrity,data synthesis,data currency  too.Scalability in this technique is quite better than the previous ones because number of interfaces does not depend on number of applications.

 Disadvantages-cost and administration.middleware products used for this is expensive than middleware products used in application integration. Database administration effort is high to implement those when it is purchased.

Middleware infrastructure integration

Middleware toolkit is used to construct a single interface which allows all applications to access all databases in this approach.This technique can vary according to complexity.A simpler interface is  provided by middleware infrastructure in simpler implementations.It implements a logical database where middleware infrastructure has intelligence and data processing abilities.

Advantages-data currency,encapsulation,data synthesis,data integrity and scalability.

Disadvantages-cost,administration.Implementing this integration needs significant administration work.Apart from  that all of the existing applications require to use the middleware architecture.These two add costs to this technique.Middleware toolkits used in this technique is expensive than those used in application.The implementation effort which is used to build middleware also costs more.So this is the most expensive technique.This technique’s performance is not good.Middleware architecture will cause some performance delay when we compare against direct access.

Migration

Migrating computer systems which  are needed to be integrated to a common database eliminates the integration problem.This can be big effort depending on the number of databases and applications involved.So it is taken into consideration when there are other reasons to use migration.

Leave a Reply

Your email address will not be published. Required fields are marked *

shares