The research of big data architecture on telecom industry

In Big Data era, telecom operators have massive data resources, such as user call data, user online data, user location data, network performance data, and so on. These data reaches PB level and is distributed in the NEs and interfaces. How to effectively carry out the collection, parsing, analysis for the amount of data to support the network construction, maintenance and optimization is a major opportunities and challenges that the telecom operators face. This paper proposes a big data platform (BDP) to solve the above problem for telecom industries. Through the collection of telecom network data, the BDP can realize data parsing, storage and analysis for MR(Measure Report), CDR(Call Detail Record), OMC(Operation and Maintenance Centre) data and etc. It can achieve a unified storage and management of all types of data. Based on BDP, operators can carry out big data analysis and data mining to realize the value of data. Finally, this paper builds a testing platform for BDP to test the performance of big data loading and big data analysis. The results of experiment show that the performance of data loading and analysis of BDP is better than traditional data warehouse. It can be applied by the telecom operators to be the foundational infrastructure to carry on the future application of big data.