Computing Software Architecture for the BM@N Experiment

Distributed data processing and computing is highly relevant to the BM@N experiment, the first ongoing experiment of the NICA project due to the high data flow, the sequential processing of which would take hundreds of years. The latest BM@N Run alone produced about half a petabyte of raw data, and when the experiment reaches its design parameters, the amount of the data will increase approximately by an order of magnitude. To solve this problem and combine all distributed resources of the experiment into a single computing and storage system as well as provide the automation of job processing flows, the computing software architecture has been developed and is being implemented. It includes a complex of software systems for distributed processing of BM@N data flows and is described in the paper.