Some Issues, Challenges and Problems of Distributed ...

1 Kamal Sheel Mishra et al, / (IJCSIT) International Journal of Computer Science and Information Technologies, Vol. 5 (4) , 2014, 4922-4925. Some Issues, Challenges and Problems of Distributed Software System Kamal Sheel Mishra#1, Anil Kumar Tripathi*2. #1*2. Deapartment of Computer Science & Engineering, Indian Institute of Technology(Banaras Hinhu University). Varanasi, India Abstract From the last two decades the trends in the Some of examples of Distributed Systems are world wide Computing Industry is towards Distributed , low cost unit, and web information, resource sharing, Clusters, Network of high unit volume product. Moreover software development workstations, Distributed manufacturing system ( , activity is becoming more decentralized thereby expanding automated assembly line), Network of branch office more development efforts for organizations. The performance of computing system has increased drastically with the computers, Information system to handle automatic inclusion of Multiprocessing and Multicomputing.

2 This paper processing of orders and Network of embedded systems presents a systematic literature review on the issues, This paper is organized as follows: Section II discusses the Challenges , Problems and solutions of the Distributed Software related work done in this area. In Section III Different System (DSS). This paper is also aimed to give a report on the issues related to DSS are discussed. Section IV shows real Problems and performance issues faced by software various Challenges faced in the way of developing DSS. professionals and researchers working on Distributed Section V discusses the Problems and solutions related to Software System. DSS so that the Challenges should be minimized. Section VI summarizes the Conclusions. Lastly Section VII gives Keywords Distributed Software, Performance evaluation, the different references used in writing this paper. task allocation, task scheduling, communication fault delay, computation fault delay, Distributed System. Distributed . SYSTEM.

3 I. INTRODUCTION. With the advent of Internet and Network technologies the Distributed Software System has become popular and important. Since the Industry is more concerned about the Distributed Distributed . Distributed Software Development it becomes essential to SOFTWARE & Distributed SOFTWARE, HARDWARE USER HARDWARE & USER. discuss the Issues related to Distributed Software System. While discussing the Issues, Challenges will come which is nothing but the constraints or restrictions on DSS. To Fig. 1 Types of Distributed System. minimize the Challenges , Problems on DSS must be discussed and the probable solution to the specific problem II. RELATED WORK. will definitely minimize the restrictions or constraints. It is This section captures the significant work done towards the accepted wisdom in the Software Engineering profession issues, Challenges and Problems of Distributed Software that developing Distributed Software System is a Systems. Many authors have identified different issues of challenging activity.

4 DSS are harder to design and program Distributed system. Sudipto Ghosh and Aditya P. Mathur[1]. because of number of processes running in parallel or described the Issues in Testing component -based process may update its variable independently or Problems Distributed systems related to concurrency , scalability, specific to the development of Distributed applications are heterogeneous platform and communication protocol. Dan not exactly implemented by known programming Nessett [2] focuses on Massively Distributed Systems: languages and software engineering environment. Design Issues and Challenges . Ahmed Khoumsi [3] worked Distributed system [Fig -1] may be categorized as i. Only on Temporal Approaches for Testing Distributed Systems. hardware and software are Distributed . ii. Users are Hiroshi Tamura, Futoshi Tasaki, Masakazu Sengoku and Distributed or iii. Both Users and hardware, softwares are Shoji Shinoda [4] focus on Scheduling Problems for a Distributed . There are several definitions on what Class of Parallel Distributed Systems.

5 Eric Koskinen and Distributed systems are. Coulouris defines a Distributed Maurice Herlihy [5] worked on Deadlocks: Efficient system as a system in which hardware or software Deadlock Detection. Ajay Kshemkalyani and Mukesh components located at networked computers communicate Singhal [6] described the Distributed Computing: and coordinate their actions only by message passing [17]; Principles, Algorithms, and Systems. S. Kartik and C. Siva and Tanenbaum defines it as A collection of independent Ram Murthy [7] worked on Improved Task-Allocation computers that appear to the users of the system as a single Algorithms to Maximize Reliability of Redundant computer [18]. Leslie Lamport said that A Distributed Distributed Computing Systems. Veljko m. milutinovic, system is one on which I cannot get any work done because jakov j. crnkovic, and catherine e. houstis [8] worked on some machine I have never heard of has crashed [16]. Simulation Study of Two Distributed Task Allocation 4922. Kamal Sheel Mishra et al, / (IJCSIT) International Journal of Computer Science and Information Technologies, Vol.

6 5 (4) , 2014, 4922-4925. Sommerville[9] described the Distributed Failure management: How can failure of the system be software Engineering. Issa Traoreand Isaac Woungang detected and repaired. worked on [10] UML-Based Performance Modeling of Distributed Software Systems. Masoud Mansouri-Samani Synchronization: One of the most important issues that and Morris Sloman[11] presented the Monitoring of engineers of Distributed systems are facing is synchronizing Distributed Systems. computations consisting of thousands of components. Current methods of synchronization like semaphores, III. ISSUES OF Distributed SOFTWRE SYSTEM monitors, barriers, remote procedure call, object method Scalability: Scaling is one of the major issues of invocation, and message passing, do not scale well. Distributed System. The scaling issue consists of dimensions like communication capacity. The system Resource identification: The resources are Distributed should be designed such that the capacity may be increased across various computers and a proper naming scheme is to with the increasing demand on the system.

7 Be designed for exact reference of the resources. Heterogeneity: It is an important design issue for the Communications: Distributed Systems have become more Distributed systems. The communications infrastructure effective with the advent of Internet but there are certain consists of channels of different capacities. End-Systems requirements for performance, reliability etc. Effective will possess a wide variety of presentation techniques. approaches to communication should be used. Objects representation and translation: Selecting the best Software Architectures: It reflects the application programming models for Distributed objects like CORBA, functionality Distributed over the logical components and Java etc. is an important issue. across the processors. Selecting the right architecture for an application is required for better quality of service. Resource management: In Distributed systems, objects consisting of resources are located on different places. Performance analysis: The Performance analysis of Routing is an issue at the network layer of the Distributed Distributed software system is a great issue.

8 It is expected system and at the application layer. Resource management that DSS should be high speed, fault tolerant and cost in a Distributed system will interact with its heterogeneous effective. It is also essential towards evaluating alternative nature. design to meet the Quality of service. [10].Moreover the ability to estimate the future performance of a large and Security and privacy: How to apply the security policies to complex Distributed software system at design time can the interdependent system is a great issue in Distributed reduce the software cost and risk. system. Since Distributed systems deal with sensitive data and information so the system must have a strong security Generating a Test data : Generating a test data to cover the and privacy measurement. Protection of Distributed system respective test criteria for testing the component is a assets, including base resources, storage, communications difficult task. It becomes more difficult in case of and user-interface I/O as well as higher-level composites of Distributed system because the number of possible paths these resources, like processes, files, messages, display increases significantly.

9 Test cases must cover the low level windows and more complex objects, are important issues in elements. Distributed system Component selection for testing: Testing Distributed Transparency: Transparency means up to what extent the components require the services of other components. Distributed system should appear to the user as a single When a component is used with other there could be a system? Distributed system must be designed to hide the possibility of deadlocks and race condition. There may be complexity of the system to a greater extent. no error detected when only one client is used because only one thread is executed but in the case of multithreading the Openness: Openness means up to what extent a system be number of clients used for testing the components may designed using standard protocols to support detect the errors. Interoperability. It is desired for developers to add new features or replace subsystem in future. To accomplish this, Test Sequence: The component is to be tested along with Distributed system must have well defined interfaces.

10 Other components. What orders should be followed in testing components? If the components do not follow the Quality of Service: How to specify the quality of service layered architectural model, there could be chances of given to system users and acceptable level of quality of cycles among the components. service delivered to the users. The quality of service is heavily dependent on the processes to be allocated to the Testing for system scalability and performance: Scalability processors in the system, resource distribution, hardware, of conventional test criteria of data is a major issue in the adaptability of the system, network etc. A good context of testing. The concept of threading may be used in performance, availability and reliability are required to the components for improved performance while testing. reflect good quality of service. But using multiple threads is a challenging task in testing. 4923. Kamal Sheel Mishra et al, / (IJCSIT) International Journal of Computer Science and Information Technologies, Vol.