Software integration means combining different applications and systems. For example, integration allows a company's ERP system to retrieve address information from a customer account management application or a company's product planning software to retrieve parts availability information from an inventory management system.
We've put together the right information package on integration for you, answering the questions: what does it mean, how should it be implemented, and what can it achieve?
Integrations are a normal part of almost any modern data system project. Even relatively small organisations tend to have several information systems in use, hence the need to develop and maintain integrations to ensure the integrity (e.g. accuracy and timeliness) of the data held by the company. In large organisations, this need is even more acute. It's therefore important to take possible integration development needs into account already at an early stage of a project.
Examples of projects
An international customer wanted to automate its production in order to reduce the workload for its operators. We planned and implemented backend services that could be used from the production equipment. The automation was started from one machine, and now most of the production machinery has been automated.
Feedback from the production operators and from customers has been positive.
A customer established a new factory and installed IT systems and integration processes. We were responsible for installing and commissioning the integration system, and were involved in the development and testing of the integration processes. As part of the project, we updated the integration system version and tested the existing processes.
The view of a developer about working with integrations highlights interactive skills and initiative. The technical solution itself can be challenging to implement. Still, typically the most time-consuming problem in integration development is figuring out from different stakeholders (business, owners, and maintainers of other information systems, etc.) what data they want to transfer, in what format, and where. The process inevitably entails many people from many different areas of specialisation, which means that finding mutual understanding takes time.
IBM App Connect Enterprise
Oracle SOA Suite
ServiceMix / Camel
Jboss + XML / XSL
becoming increasingly common
Many large companies and organisations currently have a situation in which there is a functional solution for integrations between so-called internal systems, such as an integration platform, microservices, or point-to-point integrations. Cloud solutions are increasingly being adopted, but there is no optimal single solution for integration between cloud applications and integration between cloud and internal systems.
We help our customers to choose a suitable cloud integration tool. The selection takes into account the customer's technology strategy and policies, existing technologies and software, number of integrations, volume of data transfers, scalability needs, availability requirements, fault tolerance requirements, data protection requirements, and cost structure and constraints.
One good way to ensure the suitability of the choice of cloud integration technology is to carry out a practical PoC (proof-of-concept) project. The project should choose as realistic a business case and real systems as possible to integrate with the chosen tool. At the same time, correspondence to the needs and requirements, and answers to questions that are as open as possible must be found.
How to proceed?
Usually, integration projects begin with the customer having two or more systems between which data should automatically move. There are two alternatives for implementation: either so-called point-to-point integrations, whereby the systems discuss directly with each other, or the centralised model, which uses an integration platform through which all integrations pass.
In a project, the source system(s), target system(s), and moving data are first specified. After that, we plan how data should be processed; for example, interfaces, technologies, data transfer format, possible timing issues, data conversions/compressions/enrichments, error management, and other essential issues which affect the integration in question. After planning, integration is carried out, tested, documented, and deployed. During deployment, data transfer is carried out to the necessary parties, such as to the organisation responsible for maintenance and support.
Shall we discuss more about the topic?
You are sending a message to our CEO Teemu who knows most about our customer accounts.