Until recently people in IT has been debating about various integration patterns. And EAI, AIA, WS, ESB, MOM based solutions largely solves the integration comes under:
1. Process modeling based
2. Data model based
To be on premise or cloud is new dimension. Architects, CIO etc are defining over all strategy for Cloud and on-premise, integration is missed out or only discussed at the level of feasibility(Cloud and on-premise integration). In addition there are various aspects(refer list below) which has to be considered, probably there has to be strategy for integration:).
1. Integration on-premise or cloud: Which integration should be in cloud and on-premise, how cloud and on-premise has to be integrated
2. Deployment: Where to host integration solution
3. Best practices
4. Integration cost: Every new integration adds to cost??
5. Code Portability: Same services in cloud and on-premise
6. Cloud integration skills
7. Design complexity
8. Scalability and high volume of data
9. Routing data from Cloud to On-premise or Cloud
10. Parallel processing, for Cloud to multiple targets (On-premise and/or Cloud)
11. Transformation and enrichment, when data model is not same for Cloud and target system and finaly
12. Fault tolerence
1. Process modeling based
2. Data model based
To be on premise or cloud is new dimension. Architects, CIO etc are defining over all strategy for Cloud and on-premise, integration is missed out or only discussed at the level of feasibility(Cloud and on-premise integration). In addition there are various aspects(refer list below) which has to be considered, probably there has to be strategy for integration:).
1. Integration on-premise or cloud: Which integration should be in cloud and on-premise, how cloud and on-premise has to be integrated
2. Deployment: Where to host integration solution
3. Best practices
4. Integration cost: Every new integration adds to cost??
5. Code Portability: Same services in cloud and on-premise
6. Cloud integration skills
7. Design complexity
8. Scalability and high volume of data
9. Routing data from Cloud to On-premise or Cloud
10. Parallel processing, for Cloud to multiple targets (On-premise and/or Cloud)
11. Transformation and enrichment, when data model is not same for Cloud and target system and finaly
12. Fault tolerence
No comments:
Post a Comment