支援功能模組分享之邊際運算服務配置:以穿戴式裝置為例

Fog/Edge Computing Service Provision with Function Module Sharing Enabled: A Case Study on Wearable Devices

林心鵬(Hsin-Peng Lin)

指導教授 ： 逄愛君

國立臺灣大學/電機資訊學院/資訊工程學研究所/博士(2019年)

https://doi.org/10.6342/NTU201900486

摘要

隨著感知技術和無線通信的迅速發展，物聯網（IoT）技術正在改變我們的日常生活。市場上大多數的物聯網設備本身運算及儲存資源不足，通常需要伴隨著額外的實體設備（稱之為Local-hub，大多是智慧型行動電話或電腦）透過短距離無線通訊提供物聯網設備更強大的計算資源。然而，在許多場合，攜帶實體Local-hub造成使用上的不方便；因此，很多物聯網設備亦配備了Wi-Fi或3G/LTE通訊介面，支持它們透過Internet遠程與實體Local-hub通訊。這替代解決方案稱之為Remote Local-hub，雖它可免除了攜帶實體Local-hub的不便，卻導致較長服務等待時間和較高電力消耗。為克服實體Local-hub和Remote Local-hub的缺點，本研究提出了一個Virtual Local-hub（VLH）的解決方案，它利用設備周邊的網絡設備（例如：Wi-Fi熱點或行動基地台）就近提供所需的Local-hub機能。首先，本研究將穿戴式裝置作為主要研究案例、並描述其實體Local-hub和Remote Local-hub運行機制；然後，本研究提出VLH系統架構、說明如何利用邊際計算提供Local-hub機能；接著，闡述底層VLH系統設計及實作方法，包括基於容器虛擬化和微服務技術以實現支持遠程功能模組共享的邊際計算環境；本研究更進一步探討在大規模環境下如何實現最佳化的功能模組配置和共享決策，並提出一個最佳化資源分配問題及演算法；此外，透過大規模網絡環境的模擬測試，驗證與衡量提出的資源配置演算法之有效性；最後，本研究使用智慧型手錶和樹莓派3實作一個VLH雛型系統，且根據實機量測證實VLH系統在真實世界的可實現性。

關鍵字

物聯網 ； 穿戴式裝置 ； 邊際運算 ； 微服務 ； 功能模組分享

並列摘要

With the rapid development of sensing and communication capacities, Internet of Things (IoT) technology has been changing our daily life. However, IoT devices generally require a powerful local-hub, most are smartphone or computer, to replenish processing and storage capacities for advanced functionalities. Sometime it may be inconvenient to carry a bulky local-hub; thus, many IoT devices are equipped with Wi-Fi or 3G/LTE interface, enabling them to exchange data with local-hub remotely though the Internet. However, these alternative solutions result in long response time and more power consumption. To overcome those restrictions of local-hub solutions, we present a virtual local-hub (VLH) platform, which utilizes network equipment nearby (e.g., Wi-Fi hotspot or cellular base-station) as the local-hub. In this dissertation, we first take wearable devices as the example of IoT service and describe the operating mechanism of a local-hub, then give an overview of the VLH system. After that, we depict the system design of the VLH including container-based virtualization and microservice architecture which enables remote function module sharing on the edge-computing environment. We also propose an algorithm to deal with function module allocation and sharing decisions. Finally, we verify the effectiveness and practicality of VLH via simulations under large-scale network setting and we also build a prototype system as the realization of VLH to demonstrate its feasibility.

並列關鍵字

Internet of Things (IoT) ； Wearable Devices ； Edge Computing ； Fog Computing ； Microservices ； Function Module Sharing

參考文獻

[1] M. Chernyshev, Z. Baig, O. Bello, and S. Zeadally, “Internet of Things (IoT): Research, Simulators, and Testbeds,” IEEE Internet of Things Journal, vol. 5, no. 3, pp. 1637–1647, 2018.

[2] H. El-Sayed, S. Sankar, M. Prasad, D. Puthal, A. Gupta, M. Mohanty, and C-T. Lin, “Edge of Things: The Big Picture on the Integration of Edge, IoT and the Cloud ina Distributed Computing Environment,” IEEE Access, vol. 6, pp. 1706–1717, 2018.

[3] IoT Analytics, “IoT Platforms Market Report 2018-2023,” https://iot-analytics.com/product/iot-platforms-market-report-2018-23/, 2018, [Online; accessed 12-Dec.-2018].

[4] The IoT Analytics, “IoT Platforms Vendor Comparison 2018,” https://iot-analytics.com/product/iot-platforms-vendor-comparison-2018/, 2018, [Online; accessed 12-Dec.-2018].

[5] International Data Corporation, “Worldwide Wearables Market Ticks Up 5.5%Due to Gains in Emerging Markets, Says IDC,” https://www.idc.com/getdoc.jsp?containerId=prUS44247418, 2018, [Online; accessed 12-Dec.-2018].