Magnitogorsk Iron & Steel Works (MMK) is launching an industrial audit system on the basis of big data processing technology. The enterprise reports that the ubiquitous development of IT within today's automated production process results in accumulation of enormous amounts of data. This means that further successful development is only possible if extra-big data volumes, including non-structured data, can be processed efficiently and quickly. This task can be fulfilled with the help of BigData.

MMK felt an acute need for big data processing after a set of requirements was created for information audit of various information systems used for management of the company's production processes. Each of these systems focuses on a particular task, so together they ensure the necessary product quality. The information audit inspects not the production data themselves but the changes of these data in the information systems during the production process. The volume of such information is ten-fold bigger than the end volume of the data themselves, and conventional methods can only partially offer a solution to the challenge of processing this.
MMK decided to use the most wide-spread big data processing software for this task - Apache Hadoop.

The company has already purchased the equipment necessary for the creation of a distributed system, and this equipment is currently being commissioned. At the same time, MMK Informservice experts designed the architecture of the production systems information audit and are working on software modules that will make it possible to accumulate and process the enormous amount of information from disconnected sources.