How the microservice approach to analytical decisions implementation impacts business
Sergey Kravchenko, Senior data analyst, Beltel Datanomics
Most companies today are seeking to create their own multifunctional platform for business management at different levels. Development and maintenance of such a system proves to be a challenging task, because its functionality changes as the company develops.
In the last decade, a microservice style in complex applications designing has become generally well-known. It’s the style where a single system is developed as a set of small services with each of them functioning as an independent unit and communicating with the others using standard mechanisms like the http protocol.
The Datanomics team uses the microservice approach to implement analytics systems. The main idea is to turn the customer’s platform into a single center of business processes management, including management of our services. Unlike some other companies that offer to the customer a separate software product that deals with a specific task, Datanomics suggests developing a single platform of business processes management at the customer’s sites, adding its own services responsible for specific tasks.
The main advantages revealed during implementation of analytical systems based on the microservice approach are:
- possibility to develop a unified multifunctional platform by adding independent software elements into it
- possibility to update microservice elements without a full “rebuild” of the entire system
- possibility to scale processes by isolating resource-heavy tasks demanding great computing efforts into separate modules that can function on dedicated hardware independent of the main platform; this splitting allows you to increase computing resources of service modules without interruption of the main platform functioning
- no need to implement « hostile » single-purpose software, no need to train staff
Find out more about platform business organization in the article of Anna Plemyashova, the Head of AI and analytics department.