WSO2 APIs: External Vs Internal Gateway Implementation
The article following describes two different approaches to deploying the internal and external WSO2 gateway in the AWS environment, with their pros and cons.
Join the DZone community and get the full member experience.Join For Free
Architecture Possible Approaches
1. Using different EC2 instances for both internal and external gateways:
- Process: New EC2 instances need to be created.
- To serve within the AWS network, internal DNS will be needed and LB is not required.
- Then GW service will be deployed over internal DNS and required PORT opening and listener configuration to be in place.
- Finally, we will need to request for a DNS record to be used by internal customers and record data mapping should be done.
- High availability will also be there even if other instances go down.
- No impact will be there on the internal GW setup and vice versa.
- The cost will increase by at least double.
- The entire setup needs to be done freshly which will be a time-consuming activity.
2. Using the same EC2 instance for both internal and external gateways:
- Process: A new Internal DNS is created while an LB is not needed.The
- The same WSO2 GW service, which is in use by the external GW, will be deployed over 2 separate containers in the existing EC2 instance.
- The same ports can be used, but connectivity needs to be established.
- Need to order the new DNS and record data mapping so they connect similarly.
- By utilizing the same EC2 instance, the cost can be optimized.
- Setup can be completed quite fast as compared to the other approach.
- Less availability and scalability compared to the other approach.
Published at DZone with permission of Arunkumar Patil. See the original article here.
Opinions expressed by DZone contributors are their own.