The rise of cloud computing naturally leads to the rising need for data integration from cloud-to-enterprise, and even cloud-to-cloud. However, most data integration solutions are on-premise. Thus, there is a clear need for cloud-delivered data integration, such as Informatica On Demand, which is an example of data integration-as-a-service. Let’s look into this a bit more.
Data integration-as-a-service is the ability to deliver a complete data integration stack from the cloud, including interfacing with applications, semantic mediation, flow control, logging, and integration design. In essence, data integration-as-a-service includes most of the features and functions found within traditional on-premise data integration technology, but it is delivered as-a-service over the Internet.
Data integration is a tough problem to solve, and integration on-demand does not make that any easier. The core notion is that you link up to many information systems, either at the data or behavior levels, and abstract information and/or behavior from those systems to be delivered with one or many systems, either within the same enterprise or even between companies, or more often between cloud computing providers.
There are many books on integration, including three that I have written, so we won’t get into that too much here. However, any data integration engine, on premise or in the cloud, has to support some basic functions, including:
- Semantic mediation
- Interface management
Semantic mediation means that you’re able to convert the information semantics from one system to the information semantics of another system, thus the target system is able to receive information in a format it can understand. This is core to the integration effort considering that data has specific characteristics in the native application or database, and you can account for those differences in a middle tier, versus changing each system to adhere to a common metadata model which is almost always impossible.
Interface management means that you’re able to connect into the source or target systems using whatever interface they expose. This includes database APIs, or traditional data languages.
Logging means that you’re able to log all integration activities, such as messages flowing in and out, as well as other events. This is helpful when looking at the chain-of-custody of data that supports auditing requirements, and other business requirements where the movement and changing of the data needs to be tracked.
The core benefit of data integration-as-a-service is the ability to drive the data integration function from the clouds, and thus enjoy the normal cloud computing benefits such as not having to purchase and maintain your own hardware and software. Moreover, data integration-as-a-service allows you to expand your integration capacity, as needed, without additional capital expenditure.
The largest benefit of data integration-as-a-service is that your data integration solution lives and works on the same platform, the Web, as your cloud computing provider, such as Salesforce.com. Thus, integration is just a matter of reaching across the Internet, and not through the firewall. Coupled with the cost and efficiency benefits, data integration-as-a-service provides an ongoing value that will track right along with the rise of cloud computing.