Manas has been passionate about using IT to solve business challenges for the past 18 years, majorly using TIBCO technologies. He is always searching for cutting edge, innovative technologies which can participate in real life solutions. He adopts a highly hands-on approach to technology exploration and is never too far from a computing device.
When away from a computer, he's usually out for a run or doing something cool with family-n-friends.
Solving pervasive integration challenges in Logistics with Machine Learning
With rising adoption of “consumer-focused” technologies across businesses and consumers, challenges unique to its regulations in cybersecurity and data privacy are also on the rise. As more systems are integrated across organizations and entities in different geographical locations, and maintained by various parties, integration itself, while pervasive, may become exceedingly complicated and fragmented, resulting in possible cyber vulnerabilities.
What is “holistic integration”?
Application architecture today is all about building APIs and event-driven microservices. A digital business platform will build on traditional integration technologies, extending their parameters to address emerging use cases such as microservices, serverless, and function-as-a-service architectures, as well as machine learning and edge computing.
How would you identify elements making up a holistic and pervasive integration? Let’s look at three of them in details and what they mean.
Machine Learning (ML)
These three technologies will redefine how we leverage cloud computing. The current form of cloud computing is somewhat reminiscent of the era of mainframes of the 1970s. Similar to mainframes, the computing power is centralized with clients accessing it remotely. However, this architecture is rapidly changing to enable even more efficient operation.
Instead of being forced into a centralized cloud computing model, developers now have the option to adopt a hybrid approach of both centralized and distributed architectures. Key elements, technologies, and components of cloud computing can now be implemented at individual locations at the “edge” of the cloud or disconnected from the cloud. With lightweight frameworks, a new breed of applications can leverage the same core building blocks of the cloud, such as compute, storage, and networking, on smaller, lower-resource edge devices. This model of computing, where cloud computing resources are available locally for hosting lightweight, data-driven applications, is known as edge computing.
The shift to cloud computing fundamentally changed the way software is built and consumed by developers. Along with core infrastructure, higher-level components of the technology stack started to become available as services. The rise of APIs and mobile computing prompted cloud providers to deliver Platform as a Service (PaaS) and Backend as a Service (BaaS). Developers targeting next-generation experiences through the web, mobile, and wearable applications started to consume these services by letting the cloud providers manage the mundane but critical aspects of software such as compute, storage, networking, management, security, and scaling.
Machine Learning (ML) is becoming an integral part of modern applications.  From the web to mobile to IoT, ML is powering the new breed of applications through improved natural user experiences and built-in intelligence. The availability of data, ample storage capacity, and sufficient computing power are essential for implementing Machine Learning. Cloud computing is a natural fit for dealing with several aspects of the Machine Learning pipeline.
The next generation applications need platforms that bring the power of machine learning and serverless computing to the edge. The convergence of ML, edge computing, and serverless will become the backbone of digital transformation across many industry verticals, such as, logistics, manufacturing, automobile, healthcare, finance, and insurance.
Use case: How a solution for smart scalability works for the Logistics industry
International transportations are a good candidate for edge devices, especially ones that are designed to address unique business needs, such as temperature-regulated or cold chain management. Edge devices supplied within these transportation containers can work to transmit important data related to location, integrity and status of the container. TIBCO Flogo® Enterprise is an ideal tool for this edge computing device due to its ultralight framework and capabilities for micro services build.
Incoming feed of data can be processed in the cloud as an event which can be acted upon using a server less function implemented using TIBCO Flogo. Similar serverless computing platform such as AWS Lambda, part of Amazon Web Services, is fully compatible with Flogo.
Data generated through these incoming feeds will also become valuable input for data scientists to generate their predictive models using something like TIBCO Data Science. Models can be written to forecast capacity requirements in different parts of the world, procurement needs for new containers based on aging of current ones, environmental dependencies (internal and external), and of course, being able to use these forecasting models to determine price points or develop solutions for prospective customers with similar requirements. This way, data working through a holistic and scalable solution for pervasive integration can benefit both customers and businesses - a win-win outcome for all.