An event mesh offers the connectivity, dynamic routing, scalability, high performance, and security required to run real-time logistics and operations management applications.
Most transportation and logistics operations are processes comprising a sequence of events, decisions, and actions. A regional train arrives at a depot in time for departing passengers to board a local tram or bus. The process is repeated throughout the network trains, depots, trams, and buses. Or a wholesaler holds a truck destined for a regional distribution center waiting for perishables to arrive from multiple manufacturers, whose products are on the way via rail and trucks used by private shippers.
Managing the operations requires information about an event (e.g., a train or shipment has arrived). That information is used to make a decision (e.g., have enough passengers arrived or do I have all the perishable items I need for this trip). Then an action is taken (e.g., let the local bus depart or send a loaded truck to its next destination).
The traditional approach to transportation and logistics operations that networks try to react to disruptions, but the delay from detecting disruptions to being able to act has been constrained by manual processes and limited connectivity. Such an approach is outdated. Many of these older systems use historical data (e.g., standard timetables). They do not take the many changes and disruptions (e.g., a regional flood has caused a cancelation of rail service or highways to be shut down) that occur regularly into account. Nor do they provide the agility needed in complex logistics or transportation systems.
Enter modern logistics and transportation methods. Such methods use streaming data, such as location and Internet of Things (IoT) data. Real-time operational management systems ingest and analyze such data and make an informed decision based on the real-world conditions at that moment. Increasingly, at the heart of such efforts is a peer-to-peer event routing mesh with redundant paths that can optimize information sharing in transportation and logistics applications.
Information Sharing Is Critical
Modern logistics and transportation operations require high availability and agility to adapt to changes as they happen. Furthermore, the components of the network (trains, trucks, buses, planes, etc.) are often only occasionally connected. What’s required is continuous monitoring of current conditions and the ability to react to events in real time, especially as components go in and out of coverage.
Architectures that work well in other domains do not deliver such capabilities. For example, consider a service-oriented architecture (SOA) used in typical database applications. SOA uses a classical request-reply invocation paradigm. The consumer requests a service from a service provider and receives a response. Pure service orientation with a request-reply interaction is ill-suited for today’s systems that include streams of events from different systems.
Why? In many logistics and transportation applications, data is continuously produced by sensors and embedded systems and can provide real-time information on many operational aspects. For example, RFID information read at a warehouse, the GPS coordinates of a truck, or the temperature sensor’s readings in a container with food can be combined with other data and used to derive information upon which to take actions. A simple example is to use the GPS information of a truck’s location and real-time traffic conditions to determine that a delivery will be two hours late.
The responses to such detected situations should be packaged as services. By encapsulating a given functionality as a service and providing a standardized interface, it becomes much easier to build new systems and to adapt to rapidly changing demands. However, the way these services are invoked is different for each system. Most systems based on SOA are mostly custom-tailored to the needs of a specific organization and its workflows.
Therein lies the problem. While SOAs have provided a platform for structuring services within and across enterprises, effective monitoring and timely reaction to emerging situations based on multiple sources from different parties requires an ability to integrate event processing.
The challenge comes in due to circumstances commonly found in logistics and transportation. As mentioned previously, there often are times when important components in the network are only occasionally network connected . A train that shares location and performance details may be passing through a tunnel, or a tractor-trailer drives through a stretch of highway where its cellular data carrier does not provide service.
As a result, decentralized decision-making and information flow play a crucial role in logistics and transportation. Addressing the intermittent connectivity requires a system that can avoid operational disruptions and route around unavailable nodes. The way to do that is to use a peer-to-peer event routing mesh with redundant paths to optimize information sharing.
Using an Event Mesh
The adoption of cloud and microservices for the development of modern applications makes it hard to enable event-driven interactions across distributed applications. What’s needed is a solution that supports messaging and streaming in event-driven applications.
An Event Mesh solves this problem by having nodes in a distributed mesh that automatically route around unavailable nodes. Furthermore, moving objects in the node can always get the data they need by recalculating what are the nearest nodes.
An event mesh delivers such capabilities, solving the complexity issues of distributed event-driven applications. Specifically, an event mesh offers the connectivity, dynamic routing, scalability, high performance, and security required to run real-time logistics and transportation operations management applications.
Gaining Technology Expertise
Most organizations are not familiar with developing, deploying, or managing event meshes. One way to overcome such an internal skill deficit is to team with a technology partner that offers the needed expertise.
Red Hat is a leader in event mesh. It has extensive real-world experience with EDA and service mesh technology, which takes the logic governing service-to-service communication out of individual services and abstracts it to a layer of infrastructure. Such capabilities are essential for an event mesh. Additionally, Red Hat offers open-source examples and code to help organizations with distributed mesh solutions.
To learn more about what an event-driven architecture is and how it helps with agile integration, visit: https://www.redhat.com/en/topics/integration/what-is-event-driven-architecture
