Improve business agility with a modern messaging & integration infrastructure

Challenges

Messaging is the critical part of the connectivity layer required for the exchange of data and services across a decentralized architecture. Enterprises are looking to meet the combined requirements of complexity, speed, reliability, and security to connect the digital world of applications and data. The key challenges are

Secure and Reliable Connectivity across Multiple Clouds

Applications are being developed and deployed rapidly today. However, when designing and coding it is difficult to predict where an application might run. Even with short development cycles, business strategy can change and the technology needs to support any or all of the following: Virtual Machines, Containers, Private Cloud, Public Cloud. So where does that leave the applications? Each application needs to be connected to multiple other applications to bring value to the business but in virtually every business, the deployment and running of the application can shift between different environments depending on the needs of the business. Consider the difficulty of connecting and exchanging data when you must dynamically setup the connectivity between applications that can be deployed and run anywhere. If you consider businesses that need to deploy extra capacity to meet peak workload demands, where capacity might be provisioned anywhere, then consider the difficulty of building an application to handle that level of connection flexibility.

You need to maintain the same quality of service, deliver the data reliably and securely, protect the data movement, and the application integrity in the case of failures.

Trying to do this in the application, or with synchronous links using API calls would be exceedingly complex. Building agile applications using APIs is very powerful, but APIs and stateless, synchronous connectivity don’t provide the resilience needed.

Use Case: Secure and Reliable Connectivity across Clouds for Agile Deployment

As businesses roll out applications across a hybrid infrastructure, business behavior can vary according to the location, the time of year, or even the time of day. In this example use case, an enterprise needs deployment flexibility of applications and connectivity to account for local business needs, including deployment on different clouds and complying with local laws.

To increase customer engagement and drive a higher revenue rate from customer engagement, new customer focused applications are developed to provide a more customized way to interact with clients. These new applications need to be populated with customer data from cloud-based CRM systems and connect to core business ordering, invoicing and shipping applications. Depending on global location, these systems may be public cloud-hosted, or running on-premises in Virtual Machines or, in on-premises private clouds. In most locations, these new applications may run on public clouds, but different locations may have selected different cloud providers. In some locations there are additional rules about customer and business data not leaving the country.

The new engagement applications need to be developed once, irrespective of these differing deployment options. Once deployed, and with increasing customer interaction and orders, each location need to follow a demand-driven deployment strategy to handle peaks in workload, including agile deployment of additional instances of extra capacity, whether on public cloud on on-premises.

Messaging middleware was chosen to connect the applications as they engage with other cloud-hosted solutions and on-premises deployed applications. It provided a common connectivity interface that allowed all applications to simply connect to the messaging layer without the need to be aware of which location or infrastructure they were running in, or where the destination application was running. Therefore, the business gained the benefits of transparent and fast cloud deployment, without any of the connectivity complexity.

Use case: Once and once only delivery of data for simpler applications

Many existing and new applications which are exchanging data together rely on the data being delivered once, and once only. While this might be obvious for scenarios like payment systems, it is also essential for many other systems, such as order, delivery, and billing systems. Across the board, it is critical that the data being exchanged is delivered in the order it’s sent, since you don’t want to lose an order or process a cancellation before you process the initial order. But it is also critical that the data isn’t duplicated and sent more than once. Bill your customer twice and you will have an unhappy customer.

So, what’s the solution to ensuring that your data is sent once, and once only? If you are relying on everything working, then you are likely to be disappointed, as errors and failures will happen, either in the application, the network or the systems. And if the applications are enhanced to try and check and fix delivery issues, then there will be an ever-increasing amount of error handling logic that will swamp the business logic, which will cause a significant risk of inconsistent delivery as well as increase costs.

Messaging middleware can offer a solution of once and once only delivery, removing the need for additional complex logic in the application, and even provide message ordering to ensure that messages are received and processed in the right order. These features make a big difference to the success of the business and the simplicity and focus of the applications.

Use Case: Automation Across Multiple Clouds and the Partner Ecosystem

As enterprises focus on systematically increasing their levels of automation, the combination of APIs, process orchestration, and secure and reliable messaging plays a central role. In this example use case, an enterprise has to fully automate all processes touching the shopping cart order to comply with a new two-day delivery service level to better compete in the market.

Once an order is placed, it must be processed by the organization's financial systems, updated in the SaaS marketing application, delivered to the organization's data warehouse and to the new cloud- based streaming analytics solution, and then either shipped directly from a manufacturer or fulfilled by warehouse and logistics partners.

The company previously delivered orders by batch using the EDI x12 format standard to manufacturers and logistics providers. Because the cycle time shortened, it shifted to an API-based approach using secure and reliable messaging to connect the new front-end application to the existing back-end systems of record. Fault tolerance was also required to ensure none of the systems lost any part of the transaction. Process orchestration was used with the messaging to keep track of process completion acknowledgements to ensure that each step was executed accurately.

Messaging middleware was containerized and deployed locally with each application in the datacenter, co-located with each cloud involved with the orchestration and with partners to improve speed and reliability. With the correct messaging middleware solution, the company could be sure that no orders were duplicated or dropped, that payments were processed correctly, and that delivery would occur within the appropriate window.