What's Next for the Next-Generation iPaaS?
What's Next for the Next-Generation iPaaS?
What's next in the next-generation iPaaS? Let's take a look at research happening in the integration domain and specifically iPaaS.
Join the DZone community and get the full member experience.Join For Free
The State of API Integration 2018: Get Cloud Elements’ report for the most comprehensive breakdown of the API integration industry’s past, present, and future.
Most organizations currently have medium-to-complete footprints of Cloud and SaaS applications. Once an organization tastes the cloud benefits, the pace of adoption usually increases until only the core on-premise applications remain that have no relevant cloud substitute or the “payoff” for replacing with a cloud alternative is not justifiable. SaaS (Software-as-a-Service) providers deliver newer capabilities in less time-to-market, leading to faster innovation. The SaaS products further drive the integration platform to adhere to its faster release cycles, all the while, supporting business agility.
Some may ask,"How good is a feature that is configured in SaaS within 10 min but takes a few days to integrate that feature to another app?" An often-asked complaint, but valid!
iPaaS (integration Platform-as-a-Service) addressed this challenge elegantly not only with intuitive UI-based design but also by expanding the integration platform for “self-service” development for application teams. iPaaS providers have focused on ease-of-use such that many iPaaS offerings are much more productive than classic on-premise integration platforms. This helped project managers to reuse their existing application developers to build integrations, where the integration requirements are many but the available budget and resources are limited. iPaaS is viewed as a quick solution for cloud-to-cloud and cloud-to-on-premise data integrations. iPaaS platforms continued to enhance their platform by including API management, data hubs, mobile enablement, workflow automation, B2B integration and so on, and was deemed as a “next-Generation integration technology.”
However, this was five years back! The first iPaaS entered the market almost 10 years ago and the products are more matured now. Massimo Pezzini from Gartner Inc., a global research firm, reported iPaaS as a rapidly growing market that grew almost 70% in 2017, breaking $1B. At present, iPaaS is positioned well among the mainstream systems in Enterprise Application Architecture and established its place in Hybrid Integration Platform. Organizations already reaped (and are still reaping) the benefits of iPaaS, and they are well aware of how to utilize iPaaS’s incremental releases for their digital transformation and address ever-evolving integration needs.
So now, what's next and big in the next-generation iPaaS?
Here is the exciting research happening in the integration domain, specifically iPaaS, that greatly benefits businesses:
IoT and Big Data are great opportunities that any integration provider cannot afford to ignore. In my view, it is essential to gradually offer “self-service” capabilities and simplify integrating “Things.” "Things" could generate time-series data from sensor devices, activity tracking events from wearables, wellness stream from medical devices, etc. iPaaS provides a cloud data ingestion layer where the event data is collected directly from IoT (Internet-of-Things) devices, process the streaming data and further sends to its data lake platform for core analytics. Edge analytics are applied to continuously streaming data for identifying patterns, machine failures, and anomalies. In low-latency use cases, more value is derived on edge computing the real-time streaming data at the ingestion layer rather than at downstream data lake layer.
I observed that iPaaS providers such as Informatica and SnapLogic have big data management cloud solutions for addressing high-speed mass ingestion and edge data streaming. In addition, iPaaS providers need to address the myriad connectivity issues with “Things” using varied specialized protocols, which is typically delegated to IoT gateway vendors. In the IoT area, iPaaS providers compete with PaaS providers such as PTC ThingWorx, AWS IoT, C3IoT, etc.
Widespread use of IoT and API design are encouraging companies to embrace MicroServices architecture. Breaking down a huge monolithic application stack into numerous modular microservices that are deployed on its own container improves agility, scalability, and speed with lower risk.
Microservices specific to an organization are built by its IT team and deployed on its own container execution environment - be it on IaaS, PaaS or on-premise. However, iPaaS offers common utilities and generic functionalities as reusable services. iPaaS should provide a container environment but "abstract" it for developers. It should also provide access to integration code artifacts and deployment APIs so that IT team can integrate into their own DevOps environment for deployment automations.
For example, the AnyPoint platform from MuleSoft has great support for Microservices
Machine Learning (ML) has numerous use cases in integration space that iPaaS can address:
- ML @Execution-time: Build "self-healing" integrations, identify data-quality issues, apply data-masking (conforming to regulations such as GDPR) on messages, identify patterns or anomalies on streaming data, generate predictions on transactions etc.
- ML @Design-time: Build "self-defining" integrations, auto meta-data identifications, generate auto-matching or provide recommendations for canonical models, auto type-casting, assess conformance to standards, auto code-reviews etc.
Some of the above features are already offered in iPaaS today, especially at design-time.
Artificial Intelligence (AI) expands ML use cases to a new whole level. AI-driven chatbots or voice commands, using Natural Language Processing, connects to multiple pre-defined (or self-connect at execution time) applications, extracts "data" and derive meaningful "information" that is readily usable. AI is used in developing self-defining and self-healing integrations. I believe most of the mentioned features are in continuous improvement and may be available sooner than later.
Informatica recently showcased AI-driven CLAIRE at @InformaticaWorld 2018 that builds self-defining integrations and connects to a pre-defined application based on voice commands.
Big companies such as Microsoft, IBM, SAP, and Oracle are working on to offer BlockChain-as-a-Service (BaaS). A private or consortium blockchain with distributed ledger and smart contracts have several uses cases for organizations. For example, to build a data lake for Master Data Management (MDM) providing single-version-of-truth with peer application network validations and traceability. This in turns helps in data quality and data governance, such as in GDPR. Few iPaaS provides are researching in this area to provide integration with blockchain technology. It may be a few years into the future to integrate blockchain as an iPaaS offering.
In addition to the above, there are continuous improvements happening in iPaaS such as Converge EAI & ETL functionalities, Self-service B2B (Dell Boomi has good features), Connect Everything or offer marketplace for connectors, Build-once-deploy-many (same code can be deployed on Oracle SOA Suite or SOA CS), extensive CI/CD (DevOps) supportability, support various patterns of on-premise-to-on-premise integrations and so on.
As a conclusion, due to a shift in organization’s application landscape from on-premise to cloud, the iPaaS is well positioned to addresses the next generation integration challenges and the benefits of iPaaS far outweigh the current shortcomings. With proper due-diligence, organizations will significantly benefit from the on-going research happening in iPaaS integration space.
iPaaS Providers: Informatica Cloud, Dell Boomi, Mulesoft, SnapLogic, Oracle, IBM, etc.
The author would like to thank Massimo Pezzini, Gartner, Inc. — VP and Research Fellow, for his insightful comments.
Opinions expressed by DZone contributors are their own.