To drive competitive advantage, organizations need to create a better customer experience. Organizations worldwide have been modernizing their applications to enhance this aspect by enriching their data, improving their Application Programming Interfaces (APIs) and their ability to integrate between applications in a multi-cloud or a hybrid environment. Organizations also need to integrate and orchestrate data movement between distinct applications using various mechanisms such as File Transfer, Web-Services, Message Queues, etc. Dealing with this in multiple latencies (e.g., real-time, event-driven, scheduled and on-demand) can worsen these already complex point-to point interactions to fulfill organizational business needs.
The Informatica Cloud Application Integration (CAI) service offers a single trusted solution to meet the above challenges. Enterprises can now leverage this service to support their integration requirements such as the various integration patterns, data sets, or endpoints to automate business processes, expedite transactions, and enable real-time analytics. Not only is Informatica’s CAI service the most comprehensive Integration Platform as a Service (iPaaS) solution, but it provides key benefits that include the ability to:
- Implement processes and APIs with zero code
- Build APIs with sophisticated data integration capabilities
- Leverage out-of-the-box connectivity to over 400 endpoints or develop custom connectivity in seconds
- Future-proof integration needs with a microservices architecture
- Support mission-critical apps on a high-performing, scalable, available platform supporting trillions of transactions per month
The incentive to incorporate CAI into an enterprise for a variety of use cases are many. While the overall solution can be complex, this roadmap provides guided steps to implement this next generation iPaaS service within an enterprise.
Getting Started with Cloud Application Integration
The implementation strategy for CAI is to create a “Minimum Viable Product (MVP)” by launching use cases that will drive business benefits quickly and that can be validated by stakeholders. For expediency, Informatica recommends leveraging as many out-of-the-box capabilities and templates as possible to build upon these services as the organization is navigating through the Use Case Roadmap. The approach is agile and iterative delivering incremental business value with each step along the journey.
Below are the six steps in the Application Integration roadmap.
Analyze and Evaluate the Business Requirements
An evaluation of business requirements is the first step for a successful CAI implementation. Typically, workshops are conducted with the right level of participation. This may include business unit representatives such as executives, leaders, business users, data scientists, architects, data administrators, and any other important stakeholders. A list of critical issues is collated and reviewed to prioritize the most important issues that require attention that will create the highest positive impact for the business.
Once an issue is identified and framed, a process to define the use-case to solve the problem needs to be determined. This again is a team effort where various team members meet to discuss the details for converting the business problem to a story that can be captured as a use-case. The use-case should include the problem statement, description, complications, stakeholders, assumptions, business impact, history, and examples. During this phase it is important to be open-minded with an intention to validate all opinions from the team, ensuring that all perspectives are considered. At the conclusion of this phase, there should be enough information gathered to feed into the next phase.
Design
As the design will impact the overall architecture of the enterprise CAI deployment and its supporting infrastructure it is one of the most critical phases of the CAI roadmap. Even though the CAI architecture is inherently scalable, design decisions made here determine the requirement for high availability, synchronous vs asynchronous requirements, and to an extent the initial sizing requirements for the infrastructure being built. This also determines the clustering architecture for Informatica Cloud Secure Agents and the location of the gateway (i.e., whether the organization will be utilizing the CAI gateway provided by the enterprise Informatica Intelligent Data Management Cloud (IDMC) platform).
An initial step in the design phase is to translate the business requirements captured in the use-case document into a technical document. Business analysts and the various stakeholders will need to provide their input so that the architect can work with the technical team to detail the technical requirements needed to fulfill the business requirement. Once all requirements are collected the design can be elaborated by identifying functional and non-functional requirements. This may include, but is not limited to:
- The type of connectivity (endpoints)
- Integration sequences
- Execution paths – serial or parallel path
- Events and triggering mechanisms including timers
- Error handling and notifications
- Orchestration of applications/transactions/applications across various products/services operated by different business owners
The next step involves creating a Solution Architecture Document (SAD) which will feed into the creation of a detailed Technical Solution Document. Both documents (combined with the Functional Requirements Document and the Non-Functional Requirements Document) feed into the infrastructure preparation needed to stand up the right infrastructure to support the solution. Refer to IDMC platform documents for various components/features such as:
- Opening firewalls and white-listing DNS
- Installing and configuring Secure Agents (with or without clustering)
- Security architecture for authentication and authorization
- Org and Sub-org configuration and ways to use these features
- Advanced clustering to make the infrastructure highly available between various zones provided by the cloud vendors
- Load balancing
- Disaster recovery
- SAML
- Connectors and configuration information
- Cloud API Manager
- Cloud API Portal
The key outputs from this step should have enough information to build a minimum viable product.
Build the Minimum Viable Product (MVP)
When a new platform is introduced in any organization, the primary goal (with participation from the stakeholders) is to get it right. This is often easier said than done. One method to confirm validity from a stakeholder’s perspective is the ability to try the solution in full force. An MVP just does that; and gives impacted teams an opportunity to validate their user experience before extending on the current use-case and expanding on to more use-cases. This approach provides the opportunity to develop standards and protocols and iterate and improve on the overall solution before embarking on tackling bigger and better use cases that will provide the business with more competitive advantages.
Developing standards and protocols is a key aspect in simplifying the build effort. Informatica provides various patterns and assets that can accelerate development. This includes processes, guides, app connections, service connectors, process objects, and services. Product documentation includes details for potentially enabling low-code/no-code builds with reusable components.
Test the MVP
Testing ensures that functionality is in line with stakeholder’s expectations. This process can be manual, automated, or semi-automated. Testing for application integration components should go through 3 types of testing:
- Unit testing
- Integration testing
- User-acceptance testing (UAT)
While unit testing confirms the correctness of the developed code, integration testing ensures the end-to-end orchestration and integration is precise. UAT gives the stakeholders a preview of the user experience and provides an opportunity to give feedback to the solution team to iterate upon and improve the solution.
Deploy the MVP
Deployment to production is a major step in the process where the developed code is copied to the production environment. This is where the business value can be measured. Various metrics required to measure business value must be collated and analyzed to study the performance compared to the initial outlook. Feedback from the various stakeholders is key to this process and helps in understanding the positives as well as the pain points.
Expand Application Integration to Enterprise
The entire process can now be expanded as the MVP enterprise-wide as the idea has been tested and potential stakeholders have approved it. Standards, protocols, security, privacy, and infrastructure are some of the areas that are now well-defined and that can provide the acceleration to take the Application Integration enterprise-wide. Templates, patterns, and reusable components help in reducing the budget required to deliver features and functionalities required by the enterprise.
Change Management Process
Informatica recommends aligning with the organization’s change management process as part of its Cloud Application best practices. CAI is flexible in its modularized implementation and can be incrementally productionalized as part of an Agile, Iterative, or Waterfall methodology (or a hybrid of these) as required by the organization.
As shown in the above diagram, the solution team is encouraged to work with the various stakeholders to identify the current situation, challenges, and vision of the organization; this will enable the organization to fulfil their business goals which are aligned to their organizational roadmap. It also instigates an active identification of stakeholders with participation from relevant players by identifying the terms of power, trust, and modes of interaction. The Eight-step process outlined has been inspired by John Kotter from Harvard Business School and this has been successfully adopted by many major organizations around the globe.
Summary
As organizations strive to gain competitive advantage by keeping up with new technologies, sources, applications, and types of data, Informatica CAI can act as a catalyst to accelerate this journey. This is made possible by the low-code, no-code Informatica Integration Platform as a Service (iPaaS) with various pre-built templates, patterns, and accelerators. All this is provided via a simple browser-based development environment with the complexities being taken care of by the platform in a multi-tenant environment.
With CAI, organizations can now focus on critical business needs instead of worrying about peripheral concerns. This could be the deciding factor for being a competitive leader in the global market, irrespective of area of business.