Sep 23–26, 2019

SK Telecom's 5G network monitoring and 3D visualization on streaming technologies

Jonghyok Lee (SK Telecom), Chon Yong Lee (SK Telecom)
3:45pm4:25pm Thursday, September 26, 2019
Location: 1A 15/16

Who is this presentation for?

Architect and developers for preparing real-time 3D visualization of sensor data and data engineers, data architects, and telco network operation managers preparing data analysis platform and realtime inferencing.




As the largest mobile carrier in South Korea and global technology leader of world mobile industry, SK Telecom is preparing commercial scale deployment of 5G infrastructure. One of the most important part of mobile business is a stable and reliable operation of telecommunication network and it’s importance does not changes with the evolution of technology. One of the biggest changes coming with the deployment of 5G is a transformation of telco network. In 4G infrastructure, most of the network functions are provided by NF equipments from telco hardware venders, but in 5G, most of the network functions are virtualized and running on virtual machines. As a result, the scope and target domain of monitoring system should be expanded to cover not only status and events of virtualized NF (virtualized NF) and EMS (Element Management System), but also those of infrastructures such as bare metal servers, virtual machines, and hypervisors. Applications running on multiple MEC should be monitored and analyzed centrally as well.
T-CORE, the company’s monitoring and service analytics platform, monitors several most important systems including TANGO (OSS system) and Tmap (car navigation service), and provides monitoring functionality for NFV MANO (network functions virtualization management and orchestration) system which includes hypervisors, hardware infrastructures, virtual machines, and virtualized network function applications running on virtual machines in it.
5G Infra Visualization, one of the T-CORE’s UI, visualize realtime multi-layer/multi-dc NFV infrastructure based on 3D virtualized space.
This presentation will share

  • Architecture of first version of T-CORE which was originally developed for single datacenter monitoring, software components used and how development team evaluated candidate softwares for each part of the system.
  • Step by step approaches and architectures to expand system’s coverage to application monitoring, multi datacenter scale, and virtualized 5G network infrastructure, including requirements from 5G technology and use cases, which can be applied to usual multi datacenter scale application monitoring.
  • Architecture which can be generally used for realtime monitoring and data providing hub of the company at the same time.
  • Concepts and Scenes of 5G Infra Visualization which uses 3D render engine to display 5G network’s realtime physical/service/environment data.
  • How to deploy realtime multi-layer topology-based data in 3D space and deliver operator to support mission critical service.
  • Lessons learned from more than a couple of years of operation.

and will includes following architectures

  • Architecture for collecting data from multi datacenter to central hub
  • Data collection configuration management for multi datacenter environment
  • Features and architecture for enterprise data provider and infra analytics platform simultaneously
  • Processes and architecture for the analytics platform which provides full cycle from the development of anomaly detection model to the deployment of the model on realtime streaming data
  • Architecture and tricks to visualize massive multi-layer streaming data from multi datacenter through 3D virtualized space
  • Performance and optimization skills of 5G Infra Visualization client application which uses Unity(3D render engine).
    The presentation will also introduce analytics services delivered by T-CORE data scientists which helps operator develop their own anomaly detection model including several cases.

Prerequisite knowledge

Prior experience or knowledge about building architecture for real-time processing and analytics platform. Basic knowledge of realtime frameworks and in-memory solution like apache spark, kafka, and redis. Basic knowledge of 3D visualization.

What you'll learn

* Learn how SK Telecom monitors multi datacenter scale applications and services as well as infrastructures, and how to provide stored/realtime data to other systems for further processing and analysis. * Learn what should be considered to take a role of centralized analytics platform and a standardized data platform. * Learn how SK Telecom visualize 5G infrastructure data at nationwide level as well as service traffic of particular user using 3D engine * Understand 5G network architecture and learn what’s SK Telecom’s approach for stable and reliable 5G network operation.
Photo of Jonghyok Lee

Jonghyok Lee

SK Telecom

JongHyok Lee is an architect of SK Telecom and designes T-CORE, monitoring and analytics platform of the company. He has more than 20 years of experience in data processing area, and worked at IBM before joining SK Telecom as a senior architect and led design of architecture for several enterprise-wide data processing systems of many companies in various industry.

Photo of Chon Yong Lee

Chon Yong Lee

SK Telecom

Chon Yong Lee is a project manager of SK Telecom and designes 5G Infra Visualization, telco network visualization system of the company. He has 10 years of experience in telecom network area, demonstrate 5G Infra Visualization two time in MWC.

Leave a Comment or Question

Help us make this conference the best it can be for you. Have questions you'd like this speaker to address? Suggestions for issues that deserve extra attention? Feedback that you'd like to share with the speaker and other attendees?

Join the conversation here (requires login)

Contact us

For conference registration information and customer service

For more information on community discounts and trade opportunities with O’Reilly conferences

For information on exhibiting or sponsoring a conference

Contact list

View a complete list of Strata Data Conference contacts