SOADIWA: A Service-oriented Architecture for Data Interoperability in Web Applications

The realisation of Service-Oriented Architecture (SOA) to communicate data between systems running on different platforms lack an organised framework to capture the essential elements required for successful interoperability between web applications and their services. In this work, a SOA for Data Interoperability in Web Applications (SOADIWA) was designed. The architecture of SOADIWA was based on five layers, namely Web Application Layer (WAL), Quality of Service Assurance Certifier Layer (QoSACL), Web Service Layer (WSL), Visualization Input Layer (VIL) and Visualization Output Layer (VOL). In WAL, the Service Requester (SR) initiates a request for data from the Service Provider (SP) through the QoSACL to provide appropriate website via WSL for rendering of services which must be accepted, processed and returned for a particular need in VIL. The requested data is filtered in VIL for data exploration and analysis in VOL using context-sensitive visualization techniques. The purpose of QoSACL is to check and verify the claims made by the SP about its quality of service. This enabled the SR to choose the service that satisfied its needs. The implementation comprised of Java Script, Microsoft Visual Studio 2017 and NuGet packages; while the experiment was simulated on LoadUI pro application. Standard metrics such as Optimal Performance (OP) and Phased Effort Distribution (PED) were developed to test SOADIWA. These results conformed to basic web service interoperability. The work led to the integration of a host of techniques towards the creation of a novel tool that is useful in web domain using SOA approach.

SMART CITY WEBGIS APPLICATIONS: PROOF OF WORK CONCEPT FOR HIGH-LEVEL QUALITY-OF-SERVICE ASSURANCE

<p><strong>Abstract.</strong> In the frame of smart city initiatives, map-based web applications (WebGIS) are distinguished by the complexity of client-side implementation and high load on web servers. Web-map applications provide non-typical web content usually visualized by JavaScript code. For such applications, higher-level advanced approaches to quality of service (QoS) assessment are required. In this paper, a method based on the estimation of users’ interest in a web page and the information quantity provided by a web page is introduced. In order to implement it, a proof of work (PoW) concept is applied for verification of real users. Moreover, a novel metric is introduced to calculate the information quantity provided by web pages. PoW is used for verification of real users and distinguishing them from bots. Additionally, it can be utilized for balancing of web server load. For monitoring of web pages, an image-based approach is introduced. Web pages are rendered into raster images. A number of metrics describing an image are calculated. This approach allows users to compare and track various web pages, including rich web applications providing complex WebGL content. Several web services utilize the developed solutions. Analysis of harvested data proves the effectiveness of the proposed solution.</p>