On the study of Multiple-access Edge Computing

• Main Authors: Maxim Domnitskyi, 杜民思
• Other Authors: Lin, Bao-Shuh Paul
• Format: Others
• Language: en_US
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/handle/r8mq6y

碩士 === 國立交通大學 === 電機資訊國際學程 === 107 === Multi-access Edge Computing (previously Mobile Edge Computing, MEC) is one of the key technologies in fifth-generation (5G) mobile network technology due to remarkable increase in perceived user experience by reducing the transmission latency significantly and obtaining higher bandwidth locally. Meanwhile the industry urges for moving to commodity hardware driven deployments of applications and services is becoming a prevalent topic of discussion in regards to its use in the next generation of mobile networks. The concept of commodity hardware being used for development of MEC infrastructure is interesting in it’s own right and is discussed here. Furthermore, for the various and large scale IoT applications, an application-aware MEC server can improve the network efficiency to achieve the requirements of 5G network. In the following work an overview of Multiple-access Edge Computing and surrounding technologies for simulation and emulation of current (4G) and next-generation (5G) mobile and otherwise networks is given. This work also provides a simulation that discovers benefit of using MEC in 4G networks as well as uncovers a disadvantage in using overloaded MEC solution. Lastly, the orchestration method for commodity hardware driven MEC deployment for potential video streaming application is presented based on the demand for load balancing uncovered by simulation, and conceptually tested using state-of-the-art emulation technology in form of free5GC project that is used as a functioning prototype of a 5G network with real hardware.