Bizagi provides a powerful technology for a data-level integration approach known as data Virtualization and Replication.
Using this approach, existing data structures (i.e, tables) from an external data source can be seamlessly synchronized with the data model in Bizagi.
This allows you to reuse data structures and information from systems which do not have service-oriented architecture (i.e, legacy systems), and allows you to avoid having islands of information or incurring in administration overhead when keeping information synchronized in multiple sources.
You can use all of these benefits while also having Bizagi automatically manage transactions for you (i.e, rollbacks when needed).
To learn more about how data Virtualization and Replication works, their differences and their detailed benefits, refer to Integrating data.
Data Virtualization and Replication in Bizagi PaaS
Data Virtualization and Replication in Bizagi PaaS is configured in the same way as with on-premise Bizagi projects.
Furthermore, this feature works in the same way in either set up.
However, Bizagi PaaS supports SQL Server while it also uses internet connectivity, which involves some aspects you will need to consider as described below.
1. VPN connectivity.
A Virtual private network (VPN) provides an additional degree of privacy when exchanging information over the internet, by protecting transmitted data (should it be intercepted by unauthorized attackers), while also establishing a connection between two endpoints as if these were physically wired (in terms of visibility).
When integrating an on-premise data source via data Virtualization or Replication, Bizagi PaaS will connect to it through its assigned TCP service port. Therefore, you will need a VPN to allow communication and to implement it with hardened security measures.
To learn more about a VPN and its requirements, refer to VPN setup.
Access to an external data source over the internet, inherently depends upon factors which are beyond Bizagi PaaS control, such as a higher latency in data transmission, fluctuations, interference and congestion affecting the speed of the channel, the quality of the networks used during transmission, etc.
All of these factors can affect your overall user experience, and you will, therefore, need to mitigate them, consider them during your application design, or gauge them when planning to use the data Virtualization feature in Bizagi PaaS. Please note, this is not as relevant with data Replication, as this synchronizes information in an offline manner.
We recommend that you:
1. Ensure you choose a Bizagi PaaS data center whose location offers best latency when communicating with your premises.
2. Consider using a bandwidth that optimally supports the volume of information you are expecting to synchronize.
3. Design your user interfaces according to best practices in terms of performance (e.g, by using filters, adequate number of rows and displayed columns in tables, etc), especially when these forms will be worked on in mobile devices.
For details about best practices when designing forms, please refer to http://help.bizagi.com/bpm-suite/en/index.html?what_to_avoid_in_forms.htm.
4. Evaluate beforehand (e.g. test by carrying out a PoC) whether the performance impact for integration in a cloud architecture is appropriate for the requirements of your applications.
To do this, consider variables such as the location your end users will be accessing Bizagi PaaS, the use of a network with similar characteristics as expected for a production environment, or the volume of data being exchanged with your on-premise data source, among others.
In scenarios where you determine that an integration approach does not fit your needs, then you can rely on a service-oriented integration approach from your processes.
This means invoking web services from BPMN activities or directly from buttons in the user interface, as described in Application integration.