Service Design and Distributed System Reliability in Intelligence Information System Based on Service-Oriented Architecture

Abstract

Abstract-This paper presents the model of IntelligenceInformation System (IIS) based on a Service-Oriented Architecture. In this paper we propose the new service's model, based on the Intelligence cycle and other systems which are necessary for gathering Intelligence information and data.The paper is mostly focused on the system architecture and services design as a mainstream for definition of the services. Furthermore, additional attention is dedicated on the Distributed System Reliability (DSR). I.INTRODUCTION NTELLIGENCE Information System Model gives contribution in Homeland Security and Civil MilitaryEmerging Risks assessment through the possibility of providing information in an appropriate way, by implementing pushing and pulling mechanisms into information systems, selection of data and creation of information from raw data that can be used in creating intelligence products and dissemination reports for the authorities. IIn the Intelligence Information System, which is based on SOA, are applications written in different programming language. The service design should provide interoperability between applications. It indicates that the services written in different programming languages are capable to communicate. In this connotation, processing elements (web servers, application's servers, sensors for collecting data and so on) can create distributed environment for sharing information. SOA based multi-tier approach provides legacy systems to be hooked up in a new infrastructure where the new systems and legacy systems can communicate without complexity in communication protocol (SOAP messages).In the phase of creating distributed systems, it is crucial to have metric for distributed system reliability. It provides appropriate distribution of the system's components, because introduced algorithm for distributed system reliability shows where the gaps in the systems are. Also, the designers of a system can achieve higher level of system reliability using the distributed system reliability metric. In the distributed system, each node can present service and each service can present system, subsystem or processing element. Consequently, each service in the architecture has certain value of reliability. These values of reliability are very important for system designers.The paper is organized as follows. Section 2 presents related work about the research presented in the paper. Section 3 is dedicated to architecture of the system. In the Section 3 are presented different levels of the system architecture and how they are connected. Section 4 presents service design where it is mostly focused on service interface. Section 5 demonstrates the algorithm for computing distributed system reliability and we implement GEAR as an algorithm for the system reliability. Finally, in the Section 6, concluding remarks of the paper are presented. II.RELATED WORKIn [3], the quality attributes of loose coupling and autonomy for services in the context of service-oriented architecture are given. In order for services to be in...

Distributed System Reliability in Intelligence Information System

“…1 Selected parts of this article were published under nonexclusive copyright in Position papers of the Federated Conference on Computer Science and Information Systems FedCSIS 2014 (see [Achkoski, Trajkovik 2014]).…”