A large business today might install dozens or even hundreds of applications and services spread across the local data center, a private cloud, one or more public clouds and various SaaS providers. These diverse applications and services must coexist, share data and streamline increasingly complex business workflows.
To underpin this environment, many businesses would likely have a disjointed mix of APIs, middleware and custom integration code that is tough to manage, secure and maintain. The complexity can cost the business precious IT and development resources that are better spent elsewhere.
Integration platform as a service (iPaaS) has gained significant traction in recent years to provide a common foundation for many complex and diverse business integrations. IPaaS provides an array of readily available integrations with tools for users to assemble custom integrations and data flows using scripts, visual components and traditional coding faster and more easily than individual ad hoc integration efforts. A single console facilitates integration development, deployment, monitoring, logging, security and compliance.
With iPaaS, enterprises can unite almost any type of application or service regardless of its deployment location, although there are practical limits to scalability, security and functionality. Organizations must understand the tradeoffs that iPaaS products offer and carefully evaluate iPaaS products before selecting a platform.
This article is part of
- Which also includes:
- Pros and cons of the 4 best IPaaS software options
- B2B (business-to-business)
Although the iPaaS goal of universal connectivity is still futuristic, iPaaS platforms can handle a range of integrations across an assortment of business purposes.
Perhaps iPaaS' most emblematic use is to connect applications or systems for data sharing. Enterprise application integration (EAI) techniques and APIs have long served this purpose, but iPaaS offers a single common platform. The applications, services, and data may be located in-house, in the cloud, and in almost any number of outside SaaS offerings. This kind of application integration is often referred to as automatic API generation because the iPaaS platform automatically implements the interfaces and data flows needed to establish and manage each integration.
This is a variation on application-to-application integration. A microservices architecture offers a modular means to develop and deploy applications as groups of interrelated and scalable services. Microservices typically use containers for virtualization, and microservices modules communicate and interoperate through APIs to create the overarching application. With iPaaS, a business can automatically generate and publish the APIs to support the microservices application, which reduces the need to design and build APIs and accelerates the business' ability to develop microservices applications.
Businesses increasingly rely on diverse third-party partners for access to outside services, applications and data. iPaaS has proven an effective means to handle such business-to-business (B2B) integrations, and allow disparate applications to communicate without the need to architect a common API or hand-code proprietary integrations. With proper iPaaS monitoring and management, a business can enforce security standards and audit the data flow to and from its outside business partners.
The ability to share or move data from one application or system to another sounds simple, but quickly becomes complicated. For example, one application uses data in one format and another application anticipates data in another format, but both applications need current data in real time. Traditional script-driven batch data transfers are problematic in modern business. Manually invoked scripts can be overlooked or include incorrect parameters, and any change to either application might require updates to the script which requires additional time and effort. iPaaS can help to manage data format translations and even perform data validations within real-time data synchronizations.
The increasing diversity of emerging enterprise IT, with local, remote (colocation), public cloud, and hybrid cloud platforms, imposes unique requirements for communication and data. iPaaS connects local-to-cloud, cloud-to-cloud (multi-cloud) and hybrid cloud platforms. For example, businesses that use enterprise service bus or other traditional integration schemes in-house can leave those legacy integrations in place and onboard iPaaS for local-to-cloud and cloud-to-cloud connectivity.
IoT device integration
The burgeoning internet of things proliferates sensors and actuators across retail, medical, transportation and other industries, which places enormous demands on networks and device management. Every device must be installed and integrated, often with an API, to deliver data. Higher-level processing and analytics tools must integrate with the data, typically through other APIs. More integrations may be needed to connect analytical tools to enterprise systems such as ERP platforms. Many of these integrations must operate and communicate across clouds. IoT infrastructures are a common example of iPaaS' potential -- a single integration platform to connect diverse devices, tools and APIs in a flexible and scalable way.
Big data integration
Big data projects such as business analytics and business intelligence, machine learning and AI represent advanced use cases of data integration. They depend on vast arrays of data typically stored in advanced subsystems including data lakes and cloud data warehouses. Much of that data, such as IoT data, is high-volume and time-sensitive and must be conveyed and processed with minimal latency. This advanced storage environment can involve multiple APIs within the storage subsystems, along with other APIs to connect storage to analytical and visualization or training tools. IPaaS can simultaneously handle all of the integrations within the same platform and dramatically simplify multiple complex integrations.
Multiple cloud integrations
This is an advanced use case of cloud-to-cloud (multi-cloud) integration. Consider a business that extensively uses a public cloud provider's resources and services, and must integrate those with resources and services of other clouds, such as myriad different SaaS providers. Integration of many unique data stores in many different locations and in countless different formats can prove impractical for manual or custom-developed integrations. While this is a complex and uncommon example, iPaaS can resolve such complexity as it creates, implements and manages many different integrations within the same platform.
Event stream integration
Some organizations replace monolithic applications with microservices, invoked and driven by events such as a call for service rather than running full-time. An event initiates a chain of deployment and compute processes, yields a result to the requester, and then idles or dismisses the resources invoked by the event -- this concept is also called serverless or event-driven computing. The interaction of storage, compute and workflows that invoke, complete and dismiss events is highly complex with a wide array of diverse APIs. IPaaS can adeptly handle many different integrations from a single platform, and can be a core component of an event-driven environment.
Even with a wealth of interesting and beneficial use cases, iPaaS is still an evolving technology. There is no single iPaaS design or feature set, so each iPaaS platform can offer unique features and scalability. Enterprises should carefully research, test and evaluate iPaaS platforms before they commit to any one offering. Fortunately, iPaaS can coexist with API or EAI technologies while businesses develop new integrations and discover even more use cases for iPaaS.