Now is the time of Big Data Analytics and Internet of Things (IoT). We are well past Buzz words now with clear visibility and interest from clients. One platform I would like to see integrate further to this space is “Integration Platform as a Service” (iPaaS).
While IoT and Big Data bring visibility of data to the next level, iPaaS extends that data to Enterprise systems seamlessly. With Logic Apps from Microsoft and ICS from Oracle, joining others like Dell Boomi, Red Hat etc. we have seen this space expand.
IOT - Big Data - iPaaS
It's been segregated into 3 visible layers:
Layer 1 - Device Integration
Layer 2 - Cloud Data ecosystem
Layer 3 - Integration Platform for Enterprise Systems
All of these layers play specific roles and responsibilities in the architecture landscape.
- Sensors - Detect events in a device
- Actuators - Control mechanism of a device
- IoT Gateways - Bridge between devices and Internet
- Cloud Gateways - Connects to the cloud network
- Provisioning - Help to add respective device data to the storage
- Discovery - Identify the provisioned devices for proper communication
- Streaming - Process each sets of data from devices for faster decision making
- Queuing - Publish/Subscribe pattern for persistence and also scalability
- Rules and Orchestration - This part can be either inside iPaaS or can be as a stand-alone component
- Storage - Data consolidation point from big data perspective
- iPaaS - Integration platform to build message bus or broker pattern in line to integrate with Organizational Enterprise systems
The proposed architecture extends the V's of Big data further by adding 1 more - Visibility. Today most organizations will have heterogeneous IT landscape comprised of multiple Enterprise systems. Since Big Data Analytics provided the right aspects of data back to this one, iPaaS can extend those identified ones to other systems near real time. Keeping the data as a stand-alone within the big data scope and representing it through Business Intelligence platforms is one of the traditional ways of doing things. However, I would foresee a better alignment if those can be integrated to systems to build right 'visibility' into respective systems and those systems can be more effective in future.
It has surely been an interesting journey for these platforms and concepts. While we bring these 3 worlds together, we will see more maturity in the Enterprise platform. For e.g. - If any sensors send specific outages to the Big Data ecosystem and experts then perform real-time processing with right data science concepts, how about posting specific attributes to existing enterprise systems and provide more value-add through an integrated platform.
Data flow from devices would be processed inside the big data platform (of your choice) and given a meaningful structure to the data, which can be further integrated into the on-premises system. Then, the on-premises system can leverage this data to scale up the existing processes.
Data is a critical part of all these platforms. This is the same reason that Big Data Consulting & Analytics, IoT and iPaaS are used extensively. Marriage of these 3 is a definite advantage for better business decisions and process improvements. Today, systems depend on feeds from business applications in a more online/offline way. Integrating these three platform helps existing enterprise systems to leverage the data seamlessly and effectively.
All of these platforms can be well utilized in the proposed architecture to bring enhanced benefit to the business. Process re-engineering with Platform standardization is always helpful for better decision making. If we can utilize all of these platforms with Industry standards and best practices in mind, the time of “Clarity of 'Data'” across the Enterprise landscape is upon us.
*Adapted from LinkedIn post by Nandakumar Sivaraman