Maintaining consistent data among systems

There are several ways to ensure information are consistent among systems.

Let’s examine two approaches.

You can store information in a sigle system and other systems query this first one.

Another approach is to replicate data among systems.

An important point to consider is the behaviour of the environment when a new system is added.

In the first approach there is a new system querying the provider of the data.

In the second approach the information must be delivered to one more system.

Both approaches above require care, but the second is more dangerous.

Depending of the number of control messages, the amount of data being transferred will exponentially increase.

The integration architecture can also affect the environment performance. If filters are not used, all the messages will be delivered to the new system increasing the traffic in an unnecessary way.

So, when you are designing your environment think how new systems will be added.