The most significant factor of comparison in the telecommunications industry is wireless network quality. This is especially true in South Korea, where carriers must provide seamless connectivity and mobility in wireless networks to customers even during high-capacity network packet usage, such as mobile games and streaming TV and movies.
From voice of customer (VoC) analysis, SK Telecom found that key problems resulted from not only internal server failure but also external factors, such as connectivity, compatibility, and stability with other telco service providers and content providers. But collecting and analyzing network traffic for all approaches is too hard because of the massive and fast growing volume of data.
Kyungtaak Noh and Jisung Kim offer SK Telecom’s solution—a quality management system that integrates Hadoop and big data technology—and explain how they use it to efficiently visualize and utilize big data. Kyungtaak and Jisung discuss the Hadoop-based data warehouse (DW) they built to store network data in one place and the customer experience index (CEI) they created, which integrates and quantifies all key factors related to network quality. In addition, in order to make the best use of big data in the DW, they built a data self-discovery environment based on Apache Druid as a fast OLAP engine to visualize data. It has been a great help for local network operators to analyze the cause of the problems quickly.
Kyungtaak and Jisung outline how this DW and visualization technology has helped SK Telecom make remarkable innovations in quality management. As a result of these efforts, the company has been ranked at the top in customer satisfaction for 20 consecutive years.
Kyungtaak Noh is a software manager at SK Telecom. A software architect and full stack web application developer, Kyungtaak has been working in the area of visualizing and coordinating with big data applications.
©2017, O'Reilly Media, Inc. • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • firstname.lastname@example.org