The distributed systems provide difficult polices and services and this was the reason for the wide increase in usage of these distributed systems. Across the distributed systems, connecting of various independent computers at a single place is the general process for creating a distributed architecture. These systems provide accessibility for the resources of the system such as the local memory. The communication process among these independent computers is carried using the technique called “message passing technique”.
Solving the logical problems and computational problems is the most common and significant goal of the distributed systems and the problem is divided in to separate entities and the independent computers work on solving these entities of problem by an individual user. In any type of the distributed systems the significant aspect is distribution of resources among the connected computers, the sharing of resources among the independent computers is done using three different methods and the literature review chapter includes the detailed explanation of these methods.
The problem considered in the project can be achieved in the distributed system by monitoring the individual computers and computational nodes. The problems and faults that can occur in the individual systems during the monitoring can be considered as faults and these are recorded during the monitoring of computational nodes. The faults observed in the independent systems that are caused by those computers only must be tolerated perfectly and for this the distributed systems must be implemented in a perfect approach.
Distributed computing or distributed systems supports a good range of applications and the mostly supported applications are the telecommunication networks, peer to peer networks, distributed database management, wireless sensor networks, internet systems, routing algorithms and network file systems. Achieving the main goal is the main aim of distributed systems and this is carried by combined working of several independent entities using all these applications and the performance of these systems can be estimated on the whole by considering all the systems connected in that particular network. A dynamic distributed system is an advanced version of distributed system which is arrived because of the advanced growth in the technology.
This new version system has several new features which are advanced like allocation of resources for all the independent systems on the distributed systems and dynamics sharing of data and few algorithms are used for providing these features.
This dynamic distributed systems have several problems that are already present with it and the significant issues among them are controlling the mobility of processor, tolerating the failures observed in the network, when there is change in the topology then the system is updated with a update and controlling the requests of dynamic database. Across this dynamic distributed systems the most important responsibility is passing the messages and during the communication of these messages several distributed memory objects are used and simultaneously there are many chances for the occurrence of failures in the process of distributing the “shared memory object”.
The other most common problem in the dynamic distributed systems is maintaining a duplicated “database base objects” and the energetic updates from the independent systems in the network may interrupt the process of duplication and the distributed system is made ready for the faults by the kind of communication that involves the communication in groups and logical techniques of message massing.
Paper Submitted & Written by Sathish Nagarajan