description |
---|
Discussion |
Today, SR offers an API-based data sharing mechanism. Alternate data-sharing mechanisms are being considered for large-volume data sharing across networks.
While Kafka can be used for data sharing it has the following limitations:
- It recreates partitions if the the TCP connection drops (explain)
- Accessing Kafka across networks on the Internet is a challenge as it does not run on standard ports like 80,443. This makes it hard to make it work across systems as firewall rules prevent access to non-standard ports.
So Kafka is a great option for sharing data within a network, or data center but not for sharing data across data centers.
Another option is to consider the WebSub system used by MOSIP. The consumers of WebSub connect to the same via HTTP. WebSub is being used for sharing ID data for printing of ID cards.
In OpenG2P we are considering similar system for sharing registry data for printing of social ID cards.