International Workshop on Heterogeneous Mobile Edge Computing
(HMEC 2019)
in conjunction with IEEE VTC 2019 Spring
(28 April – 1 May 2019 in Kuala Lumpur, Malaysia)
(HMEC 2019)
in conjunction with IEEE VTC 2019 Spring
(28 April – 1 May 2019 in Kuala Lumpur, Malaysia)
Noriaki Kamiyama, Professor of Fukuoka University, Japan
Analyzing Effect of Edge Computing on Reduction of Web Response Time
Modern webpages consist of many rich objects dynamically produced by servers and client terminals at diverse locations, so we face an increase in web response time. To reduce the web response time, edge computing, in which dynamic objects are generated and delivered from edge nodes, is effective. For ISPs and CDN providers, it is desirable to estimate the effect of reducing the web response time when introducing edge computing. In this talk, I show a simple model of flows acquiring web objects when browsing webpages and show a simple formula that estimates the lower bound of the reduction of web response time by edge computing. I also show an investigation of the effect of edge computing on reducing the web response time in each web category, e.g., News and Sports, on the basis of the data measured by browsing about 1,000 of the most popular webpages from 12 locations in the world. One of the main findings is that the effect of edge computing was high in all the areas except North America and Europe, and the web response time was expected to be reduced by about 1.5 to 2.5 seconds in Russia, Oceania, Japan, and South America and about 4.5 seconds in Africa. |
Ying Cui, Associate Professor of Shanghai Jiao Tong University, China
Joint optimization of communications, computation and storage resources for
Mobile/Multi Access Edge Computing Joint optimization of communications, computation and storage resources plays a critical role in designing efficient mobile/multi-access edge computing (MEC) systems. This talk focuses on joint optimal resource allocation in multi-user MEC systems with computation-intensive and latency-sensitive tasks. First, I show the importance of the optimization of task operation sequences when allowing parallel transmissions and executions for different tasks. Then, I demonstrate the significance of the optimization of caching and multicasting when tasks are related and computation results or softwares can be reused. Comprehensive mathematical models are proposed, challenging optimization problems are formulated, and optimal and suboptimal solutions are obtained using various optimization techniques. Finally, numerical results show the advantages of the proposed solutions. |