When it comes to traditional enterprise applications the nature of the APIs being used doesn’t change all that often. For that reason developers of enterprise applications tend to prefer work at a higher level of abstraction that relies on connectors rather than lower level APIs to reduce the complexity associated with building composite applications in the enterprise.
What is changing rapidly is the types of devices being used and the amount of data being accessed at any one time. The simultaneously rise of mobile computing and Big Data are changing both the amount of latency that can be tolerated when accessing remote data along with the amount of data that needs to accessed at any given time.
To address that particular issue Magic Software Enterprises now offers an in-memory data grid capability to its Magic xpi Integration Platform in addition to adding connectors for SugarCRM, Sage ERP and SYSPRO applications. In total, the Magic xpi Integration Platform now supports more than 60 application connectors, including connectors for Oracle JD Edwards EnterpriseOne, JD Edwards World, SAP, IBM Lotus Notes. Microsoft Dynamics, Microsoft SharePoint and Salesforce.
According to Magic Software Enterprises Americas CEO Regev Yativ the combination of Big Data and mobile computing is creating a situation where developers need to become a lot savvier about what data is going to be accessed most frequently by mobile computing devices that typically rely on connections with limited amounts of bandwidth to access data.
As larger amounts of memory become more accessible on next-generations processors, however, Yativ says it’s becoming a lot more feasible for developers to leverage the visual development tools that Magic Software Enterprises includes with the Magic xpi Integration Platform to define what data needs to run in-memory.
The challenge that creates for developers, however, is that they need to have a better understanding of what data is critical to end users of their applications versus the traditional approach of treating all data relatively equal regardless of its actual value to the business. Connectors are a way to simplify access to relatively stable sources of business-critical information that is relatively stable. That doesn’t eliminate the need for APIs to integrate with other applications, but connectors are a handy way to consistently access certain classes of business applications.
In the meantime, in-memory computing of all forms will prove to be a major boon to developers. But as is often the case with the coming of great power there is usually a commensurate increase in the amount of responsibility being assumed.
By Michael Vizard, ProgrammableWeb.
Click for the online version