一個在邊際運算環境中建立反應式應用之框架

A Framework for Building Reactive Applications in Edge Computing Environment

葉子樵(Tzu-Chiao Yeh)

指導教授 ： 郭斯彥

國立臺灣大學/電機資訊學院/電機工程學研究所/碩士(2019年)

https://doi.org/10.6342/NTU201901397

摘要

邊際運算為新興的資訊科技，提供網路邊際端上的運算與儲存資源，使得物聯網應用具備即時控制及資料處理的能力。由於物聯網應用常以感測、控制及動作的交互式型態所建立，邊際運算軟體所編寫的程式邏輯往往基於事件驅動的結構上以解耦軟體複雜的控制邏輯， 除此之外，具備高度分散性的邊際運算節點以及可靠度較低的網路環境，亦需以事件驅動的方式反應系統狀態，以提高擴展性、可用性及反應性。反應式編程是以資料流模型組合事件驅動程式的編程模型，能有效提高程式組合性、模組化以及處理事件的關係語意。然而，目前在邊際運算上仍缺乏相關的研究使得反應式編程模型於分散式邊際運算系統上的議題值得去探討。因此，本研究之目標為探討邊際運 算上的分散式反應編程，包含提出一個分散式反應編程框架的設計與實作、評估在邊際環境下物聯網事件驅動應用的設計方法及其效能特徵。

關鍵字

邊際運算 ； 函數式編程 ； 分散式反應編程 ； 事件驅動系統 ； 軟體架構

並列摘要

Edge computing is a novel IT technology to provide computation and storage resources at network edge for real-time control and data processing capabilities in IoT application. Because most IoT applications are built in interactive style for sensing-control-action loop, event-driven programming is useful to decouple complex software. Besides, due to the characteristics of high distribution and unreliable network environment in edge computing, event-driven architecture is also useful to build a scalable, available and responsive distributed edge computing system. Reactive programming is a variant of dataflow programming model for composing event-driven program in a modularized structure with capability for semantically analyzing event relations. However, it is still lacking of reactive programming studies on distributed edge computing systems so that is valuable to investigate the design and implementation issues. Therefore, the objective of this thesis is to investigate distributed reactive programming on edge computing, including to propose a distributed reactive programming framework, and evaluate design methodology and performance characteristics for edge-based and event-driven IoT applications.

並列關鍵字

Edge Computing ； Functional Programming ； Distributed Reactive Programming ； Event-driven System ； Software Architecture

參考文獻

[1] ReactiveX. http://reactivex.io, 2019. [Online; accessed 11-March-2019].

[2] M. S. Ardekani, R. P. Singh, N. Agrawal, D. B. Terry, and R. O. Suminto. Rivulet: A fault-tolerant platform for smart-home applications. In Proceedings of the 18th ACM/IFIP/USENIX Middleware Conference, Middleware ’17, pages 41–54, New York, NY, USA, 2017. ACM.

[3] E. Bainomugisha, A. L. Carreton, T. v. Cutsem, S. Mostinckx, and W. d. Meuter. A survey on reactive programming. ACM Comput. Surv., 45(4):52:1–52:34, Aug. 2013.

[4] E. Bertoluzzo. The essence of reactive programming: A theoretical approach. Master’s thesis, Delft University of Technology, The Netherlands, 2017.

[5] C. C. Byers and P. Wetterwald. Fog computing distributing data and intelligence for resiliency and scale necessary for iot: The internet of things (ubiquity symposium). Ubiquity, 2015(November):4:1–4:12, Nov. 2015.