Distributed Model for Collaborative Computing Farzad Amirjavid 1 , Hamid Mcheick 1 1 Département d’informatique et de mathématique, Université du Québec à Chicoutimi, Canada Abstract - Since the advent of powerful mobile devices and their systems, the limitation of existing architecture model and required resources define the major problem of such devices. Increasing the computational requirements of these devices may cause certain tasks to fail under certain constraints. Collaborative computing can solve this problem. This paper proposes a new architecture model for mobile devices based on Services Oriented Architecture (SOA). This model aims at managing and optimizing distributed computing tasks such as reduce time and computational power in collaborative computing environments. Keywords: Collaborative Computing; Service-Oriented Computing; Mobile devices . 1 Introduction Since the advent of more powerful mobile devices and their technologies (e.g., Bluetooth, IEEE 802.11), architecture has become the bottleneck for developing innovation applications. The limitation of existing architecture model and required resources (storage, battery life and computational power) define the major problem of such devices [4], [5]. Increasing the computational requirements of these devices may cause certain tasks to fail under some constraints (e.g., preserving time). Further, increasing the capacity of the mobile devices doesn’t change the capacity of the servers to which they connect. For example, in the case of increasing process requests in network, servers cannot operate as fast as normal state. This may make overhead or overload the network and the servers. One of the solutions is to dispatch some work to existing computers in the network. Therefore, collaborative computing can solve this problem. Collaboration can be defined to be “the act of working together in a joint intellectual effort” [6] and [7]. The main idea here is to address the lack of communication between individuals by building a collaborative network. For shake of explanations, let us consider a person as a performer tasks, that person cannot achieve all his tasks by himself until beyond its death, regularly asking for, offering and receiving services. Maybe others perform some tasks better or faster than him. Also, sometimes he does not find time to do these tasks by himself. To solve this problem he should do his jobs faster to save time or he should use assistance by others. For example, sometimes he does not cook and instead he receives services by calling restaurants. Existing approaches such as Collaborative Computing, and middleware frameworks such as JBoss, and Websphere are all based on client-server and multi-layers in their architecture to satisfy customer requirements such as efficiency (time), computation, transparency and so on. This architecture still suffers from efficiency (by executing a request by one server), scalability, and coupling of components across the network and therefore it is difficult to reuse them when it is necessary. Developing information technology, decreasing costs and increasing the demand for information services and also considering high bandwidth demand especially for multimedia services, we have introduced a new architecture model for collaborative computing. This model can offer a better quality of service to clients, as well as provide higher security approach and scalability. Sending SMS, configuring mobile phone using personal computer, managing mobile phone contacts, file sharing and in any other services in the client–server format can be done in our proposed SOA based model. The aim of this paper is to introduce a framework that provides collaboration among devices, as well as between devices and their servers. In this model we introduce a new architecture to implement distributed computing of processes in the web, based on Service Oriented Architecture (SOA) concepts [1]. We break each task based on its technical properties into some group of codes, then assign services to each service provider using messages. Services communicate with each other (by sending and receiving message) via their interfaces. The next section includes a brief overview of the background. Distributed Computing Model is given in section3 in which we describe its implementation and its services. Section 4 describes communication issues between different parts. Then, section 5 discusses distributed computing implementation. We describe the advantages of Distributed Computing Model in section 6. We conclude in section 7. 2 Background Suppose we want to offer a service in the web like a news or weblog website. This website is so popular and so many people refer to this website daily. There are many ways to provide this service, but the simplest way to implement this idea is to use a single computer to show today’s news, however many users would like to perform search some words and request other additionally services. Consider this computer will search its database, find related images and sometimes needs to check access permissions and etc. This way we will be able to guaranty our quality of service (QOS) for a specific number of users just because available resources are restricted on our single computer and we have not predicted or assumed a scalable design.