針對基於REST風格的手機應用程式減少流量的查詢最佳化

Abstract

這幾年隨著智慧型手機的普及化，使得手機不單單只是提供通話的基本工具了，使用者可進一步透過各種手機應用程式來取得所需資訊。而這些手機應用程式大多都利用RESTful APIs發出查詢來和應用程式伺服器進行溝通的動作。但隨著使用行動網路的用戶數量不斷地攀高，硬體設施逐漸不敷使用，導致網路傳輸速度低落、頻寬不足的問題發生。傳統的解決方法是建基地台以增加頻寬，但成本非常高，不符合經濟效益。故應該從另一方面下手，試著降低對頻寬的需求。

有鑑於此，本篇論文將提出一種可降低多餘資料傳輸量的方法，使得手機應用程式在利用RESTful APIs向應用程式伺服器發出查詢前，能夠先進行一連串的判斷和處理動作，以避免向應用程式伺服器發出無意義或重複的查詢，進而達到查詢最佳化的目的。而在確保每次都發出最佳化查詢後，手機應用程式就只需向應用程式伺服器下載所欠缺的資料，即可減少不必要的資料傳輸量，以舒緩行動網路壅塞的問題。

With the rapid popularity and vast distribution of smart phones in recent years, mobile phones provide not merely basic communication but also highly customized information via their various mobile applications. While most of these mobile applications make their queries through RESTful APIs to interact with application servers, the current infrastructure can hardly satisfy all of their demands due to the exploding number of mobile subscribers, inducing network speed with relatively insufficient bandwidth.

The traditional solution to such misery is to establish new base stations directly, however unlikely, this approach sounds to be only acceptable in short terms, but impractical for permanence due to the inevitably high cost of hardware and setup. A better way to solve this problem is to reduce bandwidth demand.

In this research, we propose a mechanism to eliminate redundant data transmission before the mobile applications make queries through RESTful APIs by query optimization. After query optimization is completed, all of the mobile applications only need to download missing data from application servers, eliminating unnecessary data transmission and reducing network congestion.