locked
Azure Service Fabric and Microservices RRS feed

  • Question

  • Read some extensive documentation but still could not get the clear answer how service fabric help to simplify micro service development and deployment.

    Let’s say two services name A and B. In micro service design each service has its own database. If something changed in service A’s state, based on that change, something has to happen on service’s B state.

    Service should not communicate each other over HTTP (if one service fail other will be effected if they all are communicating over rest/http), so then eventual consistency comes where async message base communication happen (queue, event grid or service bus) between Service A and service B. This solve some issues but there is a lot of complexity involved.

    I could not find any clear direction how to achieve this in service fabric. I assume: Since in service fabric all the nodes are on private/internal/local network, there is no network latency and also services fabric node are very reliable, so services can communicate each other over http. Having each service communicate other over http omit the need for async event base communication between service.

    Is this correct? If not please help to know, how service fabric simplify microservices development and deploument.

    Monday, August 31, 2020 8:09 PM