Comparison Of Distributed, Mobile Distributed And Non-Distributed Systems

Definition of Distributed Systems

1.Differentiate between Distributed systems, mobile distributed systems and Non-distributed systems?

2.Definition and explanation of theoretical aspects of communication in distributed systems?

3.Compare and Contrast the fault-tolerance requirements in distributed system and mobile distributed system?

4.Main social issues arising from the use of ubiquitous distributed systems?

The report mainly focuses on the comparison of distributed and non-distributed system. According to Murphy et al., (2014), distributed systems are defined as a piece of software, which enables  collection of different independent computers that appear as a single coherent system for its users: whereas,  in non- distributed system all the processing are completed by  using a single processor with the help of single processing site.

The assignment discusses the theoretical aspects of communication in distributed systems and the characteristics of interprocess communication. The report illustrates the social issues related with the use of ubiquitous distributed system, which increases the cyber security threats. The report also provides number of methods for resolving the issues associated with the distributed system.

According to Kavousi-FardandSamet (2013), distributed system are defined as software in which various components are situated at networked computers for communicating as well as coordinating their actions only by passing messages. The distributed systems have some specific characteristics, which includes concurrency of components, independent failures of components and lack of global clock (Kopetz 2014).  Distributed systems help in sharing various resources and capabilities for providing users single as well as integrated coherent network. Some examples of modern distributed applications include multiplayer online games, web search and financial trading systems.

The table below reflects the difference between the distributed system and mobile distributed system

Ubiquitous distributed system contains the idea of embedding various computational powers into the daily environment in such a way that it helps in adding functionality to the environment in a very useful as well as unobtrusive manner. It is stated by Bott (2012), that it helps in harnessing very small, cheap computational devices, which are located in the physical environment of the users including home and office (Goswami & Kundu, 2013, p.29). The device that utilizes the system must have constant availability, and as a result, they are connected completely. It mainly focuses on learning by eliminating the different complexities of computing.  The main key features of ubiquitous distributed system include use of large umber expensive processors, capturing of various real time attributes and use of converging network.

Difference between Distributed Systems and Mobile Distributed Systems

The table below illustrates the factors that differentiate Ubiquitous system from non-distributed system.

According to De Palma et al., (2012), Interprocess Communication is a method, using which two or more processes communicate for exchanging data by synchronizing their activities. There are different types of interprocess communication available, which are as follows:

Local interprocess Communication:  It helps in allowing one or more processes to communicate with one another using a single system. The local interprocess communication includes:

Pipe: It is stated by that David et al., (2014) that pipe helps in passing information from one process to another process. The pipes are categorized into two types, which include named pipes and unnamed pipes.

Signals: Signals are defined as an interrupt, which are software generated. They are generally sent to the process when an event occurs.

Message Queuing: Ruiz-López et al., (2013) stated that, message queuing helps in allowing processes by utilizing a message queue

Shared Memory:  It assists sharing of memory segments among various processes for having proper communication in order to exchange data between them.

Sockets: They are considered as the end for having proper communication between different processes (Barcelos, Rigo & Barbosa, 2015). It helps in allowing communication between various program of the client and a server.

Remote interprocess communication: Qiu et al., (2013) opined that remote interprocess communication helps in allowing more than two procedures for communication among each other on different systems. Different types of networking protocol are implemented in various remote interprocess communications. (Rajguru & Apte, 2012, p.2279). The interprocess communication utilizes:

• Internet-domain-specific- socket
• The Transport Layer Interface (TLI)
• Remote Procedure Calls (RPCs).

The characteristics of interprocess communication include:

Message Passing: According to Kavousi-Fardand Samet 2013), the message, which is passed between two processes, is supported by two-message communication operations, which includes send and receive.

Synchronous form of communication:  In this form of communication, send as well as receive are considered as blocking operations.

Asynchronous form of communication: In asynchronous form, the send options is considered as non- blocking and receive operation is considered as blocking operations.

Fixed Internet Address: VermeulenandRath (2013) explain that fixed internet addresses are utilized in order to avoid one between the two approaches for providing better location transparency. The approaches include client program referral to various services by using the name and the operating system.

Reliable Communication:  It helps in providing validity as well as integrity. According to Deng et al., (2015, p.720), for achieving proper validity, point to point message service must be required for describing the message as reliable when they are guaranteed for delivering. On the other hand, Kekatos & Giannakis, (2013, p.1618) claim that integrity can be achieved if the messages have arrived uncorrupted as well as without duplication.

Ubiquitous Distributed System

Interprocess communication is very much important in the programming environment of Java for addressing different issues. RMI and COBRA are the two approaches that helps in allowing exchange of various objects for dynamically invoking various methods as well as procedures at runtime. Java helps in providing support in application development by using the different interprocess primitive (Kavousi-Fard & Samet, 2013), However, the RMI, as well as COBRA cannot be used under some of the circumstances and therefore in that situation, plain old sockets can be used for communicating between various applications of Java. Objects are serialized as well as transmitted over sockets by using two different classes including ObjectInputStreamand ObjectOutputStream (Barcelos, Rigo & Barbosa, 2015).Sockets are much simpler than RMI or COBRA and therefore one can define everything by utilizing the communication protocol. Lock files can also be utilized for communication. They are considered as primitive methods of communication between various procedures on the same system. Lock files are quite simple for communicating between two or more processes for the well-known file or file system. This is such a primitive approach, that it is not considered as a legitimate form of interprocess communication

In the distributed system, both the process, as well as communication channel, may fail. The failure model of distributed system helps in defining the ways or the procedures of failures in order to provide proper and effective understanding of failure. The failure model of distributed system is categorized as omission failure, arbitrary failure and timing failure.  According to David et al., (2014), omission failures occur when the process or the communication channel of distributed system fails in performing the actions, which it supposed to do. On the other hand, Bott (2012) stated that arbitrary failure occurs when there is any type of error present within a process and the timing failures are present as well as applicable in distributed system for liming the execution time.

On the other hand, it is stated by Vermeulen and Rath (2013), that typical failure in the process, as well as communication of mobile distributed system include node failures, loss of packets and connectivity failures. The connectivity failure mainly arises when the wireless network is unable to get connected with the network. Connectivity failures in MDS are due to unavailability of connection to the various access points. It helps in identifying other failures, including partitioning as well as isolation failures.

Mobile Distributed System has a number of additional failures as compared to distributed system. The distributed system have three types of failure modes, which includes omission failure, arbitrary failure and timing failure; whereas the mobile distributed system have three different modes of failure which include node failure, connectivity failure and loss of packets due to very weak wireless links (Francalanza, Gauci & Pace 2013)

How Ubiquitous System is Different from Non-Distributed System

According to Ruiz-López et al., (2013, p.26), the performance of the system is affected due to the failure that occurs. Some of the fault in the distributed system is narrowed down due to failure in the hardware as well as software. The errors, which have occurred in the distributed system, are mainly separated using some of the components such as omission, performance, timing as well as crash. On the other hand, Vermeulen & Rath (2013) stated that in mobile distributed system failures are dependent on the tolerance of connectivity, access point and node failures.

The extended functionalities which are offered by ubiquitous computing from cyber security raises number of social issues which include:

Privacy issues: According to De Palma et al. (2012, p.331), privacy has become one of the perverse tasks due to physical outreach of various ubiquitous computing. The computing capabilities, as well as extra intelligent spaces, are supplied mainly with the help of natural construction. The spaces are used for context information and therefore the system helps in forming a distributive observation system, which has the ability to capture information about users and thus donates confidence for tracking prevention of different users (Francalanza, Gauci & Pace 2013).

Trust issue: Due to the loss of information or due to any other privacy issue, the trust which is an association between two entities weakens (Sambasivan et al., 2014).The privacy issue rises due to de-Anonymization attack or due to leakage of information through poor setting or by any other method. The leakage of information reduces trust and it creates lot of issues.

User interaction issue:  According to Francalanza, GauciandPace (2013, p. 187) the nature of group interaction is not possible in the space possible due to absence of proper information from the users of the system. As a result the users face problems in communication or interacting properly. It is quite important to overcome this type of issue.

There are some measures which help in reducing the social issues:

Platform Safety: According to Bott (2012), Ubiquitous networks have a capability to download different applications, which allow proportional update as well as reconfiguration. Therefore, limitations from downloaded source must be present so that the malicious information cannot enter or reconfigure different instruments. Therefore, it is quite important to protect the platform from various attacks.

Single Sign on: It is very much important to have single sign in ubiquitous system. It helps in reforming the initiation of various entries by authenticating all the network domains for including different leaving as well as joining of ubiquitous network without having any disturbance.

Type and Characteristics of Interprocess Communication

Content safety: It is the opinion of Murphy et al., (2014), that the ubiquitous system helps in delivering various capabilities of different services. Therefore, the assurance of being safe as well as protected for the providers in the digital environment is provided using the system of Digital right Management.

Conclusion

 It can be concluded that distributed system as well as non-distributed system differs from each other in various aspects. Distributedsystem helps in supporting various types of clients by providing a shared middle tier; whereas,non-distributed system helps in supporting the simplest architecture of different web applications.  It has been identified that there are number of risks associated with ubiquitous distributed systems, which includes privacy issues, trust issues as well as user interaction issues. The problems or challenges can be resolved if proper methodology for application development can be followed. The report discusses various measures including platform safety, content safety as well as single sign on for reducing the impact of social issues due to the use of ubiquitous distributed system.

References

Barcelos, A, Rigo, S.J& Barbosa, J.L 2015, June, Ubiquitous System for Stroke Monitoring and Alert, In 2015 IEEE 28th International Symposium on Computer-Based Medical Systems, pp. 330-333.

Bott, R 2012, Intelligent placement and delivery of mobile advertisements and electronic coupons via a distributed system in a mobile network, U.S. Patent Application 13/572,582.

Chandhoke, S, Mohrmann, L.E, Ullrich, A.C&Greenstreet, R.D National Instruments Corporation 2013, Clock distribution in a distributed system with multiple clock domains over a switched fabric, U.S. Patent 8,583,957.

David, K, Geihs, K, Leimeister, J.M, Roßnagel, A, Schmidt, L, Stumme, G&Wacker, A 2014, Socio-technical Design of Ubiquitous Computing Systems, Springer, pp.353.

De Palma, N, Hagimont, D, Boyer, F&Broto, L 2012, Self-protection in a clustered distributed system, IEEE Transactions on Parallel and Distributed Systems, 23(2), pp.330-336.

Deng, Q, Nie, R.S, Jia, Y.L, Huang, X.Y, Li, J.M. & Li, H.K 2015, A new analytical model for non-uniformly distributed multi-fractured system in shale gas reservoirs, Journal of Natural Gas Science and Engineering, 27, pp.719-737.

Francalanza, A, Gauci, A&Pace, G.J 2013, Distributed system contract monitoring, J. Log.Algebr. Program., 82(5-7), pp.186-215.

Goswami, S &Kundu, C 2013, XML based advanced distributed database: implemented on library system. International Journal of Information Management, 33(1), pp.28-31.

Kavousi-Fard, A &Samet, H 2013, Multi-objective performance management of the capacitor allocation problem in distributed system based on adaptive modified honey bee mating optimization evolutionary algorithm, Electric Power Components and Systems, 41(13), pp.1223-1247.

Kavousi-Fard, A &Samet, H 2013, Multi-objective performance management of the capacitor allocation problem in distributed system based on adaptive modified honey bee mating optimization evolutionary algorithm, Electric Power Components and Systems, 41(13), pp.1223-1247.

Kekatos, V &Giannakis, G.B 2013, Distributed robust power system state estimation, IEEE Transactions on Power Systems, 28(2), pp.1617-1626.

Kopetz, H 2014, A Communication infrastructure for a fault tolerant distributed real-time System, Distributed Computer Control Systems 1994, p.13.

Murphy, B.T, Scheifler, R.W, Pan, Z, Waldo, J.H,Wollrath, A.M & Arnold, K.C Oracle America, Inc, 2014, Dynamic lookup service in a distributed system,U.S. Patent 8,713,089.

Ostroukh, A &Pomazanov, A 2014,Realtime Development and Testing of Distributed Data Processing System for Industrial Company, Middle East Journal of Scientific Research, 20(12), pp.2184-2193.

Qiu, M, Zhang, L, Ming, Z. Chen, Z, Qin, X. & Yang, L.T 2013, Security-aware optimization for ubiquitous computing systems with SEAT graph approach, Journal of Computer and System Sciences, 79(5), pp.518-529.

Rajguru, A. A &Apte, S.S, 2012, A comparative performance analysis of load balancing algorithms in distributed system using qualitative parameters, International Journal of Recent Technology and Engineering (IJRTE) ISSN, 1(3), pp.2277-3878.

Ruiz-López, T, Garrido, J.L, Supakkul, S & Chung, L, 2013, A Pattern Approach to Dealing with NFRs in Ubiquitous Systems, In CAiSE Forum pp. 25-32.

Sambasivan, R.R, Fonseca, R, Shafer, I& Ganger, G.R 2014, So, you want to trace your distributed system? Key design insights from years of practical experience (Vol. 102), Technical Report, CMU-PDL-14.

Vermeulen, A.H &Rath, T.A, Amazon Technologies, Inc. 2013, System and method for checkpointing state in a distributed system, U.S. Patent 8,458,517.

Wu, G, Zhang, H, Qiu, M, Ming, Z, Li, J& Qin, X 2013, A decentralized approach for mining event correlations in distributed system monitoring, Journal of parallel and Distributed Computing, 73(3), pp.330-340.

Order an Essay Now & Get These Features For Free:

Turnitin Report

Formatting

Title Page

Citation

Outline

Place an Order