FAULT TOLERANT SYSTEM USING

Document Sample
FAULT TOLERANT SYSTEM USING Powered By Docstoc
					National Conference on Role of Cloud Computing Environment in Green Communication 2012                                               406




                  FAULT TOLERANT SYSTEM USING
                 RECOVERY BLOCK MODEL FOR WEB
                            SERVICES

                                                  Jenita Amala Bai U#1, Anitha J*2
                                  #1
                                       II PG Student, Rajalakshmi Engineering college, Thandalam,
                                                       Chennai, Tamilnadu, India.
                                                E-mail:sharewithjeni@gmail.com
                                 *2
                                      Senior Lecturer, Rajalakshmi Engineering college, Thandalam ,
                                                       Chennai, Tamilnadu, India.

      Abstract— A Web Service is a unit of managed code that can be remotely invoked using Hyper Text Transfer Protocol
      (HTTP) over Simple Object Access Protocol (SOAP). The deviations of web service behavioural from the requirements
      are monitored using a Mill model. In this paper a fault tolerance approach using Recovery Block (RB) model for web
      service requirements is addressed. The aim of this proposal is to provide fault-tolerant software service even under the
      failure of some services. The RB model applies the acceptance test to the output of an individual alternate; this acceptance
      test is necessity for every distinct service. The result shows that providing web services with excellence fault tolerance
      capability.

      Key Words—Web services, Requirements, Monitoring, Fault Tolerance, Recovery Block Model.

                      I.          INTRODUCTION
          A Web service is a communication method between two electronic devices over a network. It is just like a
      software system which is designed to support interoperable machine-to-machine interaction over a network. All
      other systems interact with the web service by using SOAP messages, it can be conveyed using HTTP with an XML
      serialization in conjunction with other Web-related standards. XML and HTTP are the platform of Web services.
      XML language can be used between different platforms and programming languages and it still express complex
      messages and functions. The HTTP protocol is mostly used Internet protocol.
          While this web service technology is widely expected to enable the interoperation of heterogeneous systems and
      the reuse of distributed functions, the industry uptake of this technology has been slow [1]. Some research has
      revealed that the lack of quality assurance and guarantee, thus, leading to deviation of



      service behaviour from requirements, is one of the important factors[2].
           In some environment, software requirements are about the expected behaviour of the target software. For
      service oriented software, the behaviour of service is not only determined by the program itself, but also affected by
      many other factors [3]. It may be limited hardware resources, long response time due to low network bandwidth, low
      performance due to too many client requests and denial of service or wrong result due to malicious requests. All
      these unexpected results are deviations of Web service behaviour from the requirements. Monitoring is the first step
      to detect the faults, where the further actions such as decision making and fault tolerance are all based on the
      information extracted through monitoring.
          Fault Tolerance is the property which is used to enables a system to continue operating properly even under the
      failure of some of its components. In this paper, RB model is used which can provide fault tolerant service.

                    II.       EXISTING SYSTEM


 Department of CSE, Sun College of Engineering and Technology
National Conference on Role of Cloud Computing Environment in Green Communication 2012                                               407


          Software fault tolerance, it is just avoiding the failure service. In [4], [5], [6] papers the fault tolerant techniques
      are used to provide fault tolerant service even failure of some services.
          In [3], monitoring framework is used, to monitoring the entire process of the software service and to detect the
      faults. An monitoring framework has some components such as distributed probes, an agent, a central analyser, and
      a management centre. Different mechanisms are available to implement the distributed probes. These mechanisms
      can be classified along two dimensions: instrumentation-code based versus interceptor based.

          The instrumentation code is inserted manually by the programmers and the interceptor’s code can be inserted
      freely into any location of the target code. Interceptor can obtain details of the messages to and from the monitored
      software service. An agent acts as an active events collector in that software service. The central analyser can be
      responsible for Constraint violation detection and Problem determination of the service. From the Management
      centre, administrators of Web service can be aware of its states through human-friendly management interface and
      perform potential actions to keep the behaviours of service consistent with web service requirement [3].
          In [4], local and global fault recovery mechanisms are used to provide fault tolerant service. The local recovery
      mechanism is restricted to recover the individual service instance, while global recovery mechanism applies to the
      entire application. The local recovery mechanism tries to recover the failure service with minimal or no intervention
      by the global recovery mechanism. A local recovery mechanism sends the failure notification to the global recovery
      mechanism in case of its failure to recover the fault service locally. The overall framework provides the capability
      to,
      1. Detect a fault or failure of the software service
      2. Estimate the damage caused and decides on the strategy for recovery,
      3. Repair a fault of the software service, and
      4. Restore the application state [4].
          This framework demonstrates a method of effectively decoupling the local and global fault recovery
      mechanisms. It provides a capability for monitoring the individual service instances as well the service hosts for
      detecting the fault or failure service. Estimate the fault service caused based on that decides on which strategy going
      to be use for recovery. It can be either local or global fault recovery mechanism which is used for providing fault
      tolerant service.
          In [5], FTWeb infrastructure used for providing fault tolerance on web services. FT-Web supports the use of an
      active replication technique in order to obtain fault tolerance service in service oriented architectures. The main
      objective of this approach is to provide fault tolerant service in the following kind of faults: stop, omission and
      value. The service objects that can able to provide the basic functionalities for building fault tolerant distributed
      applications are Replication Management Service (RMS), Fault Management Service and Logging and Recovery
      Service.
          The RMS interacts directly with the Object Group Management Service, acting dynamically in the input and
      output of replicated objects. The Fault Management Service performs the interfaces of the fault monitoring and
      notification services. The Fault Notification Service performs the function of informing RMS of the faults recorded
      by the detectors. The main objective of the Recovery and Logging Service is registering requests received by the
      server, keep the state of the replicas consistent and carry out
      recovery procedures on faulty replicas [5].
          This paper presented a proposal for deployment of the active replication technique in order to achieve fault
      tolerance service. This approach provides tolerance service only for stop, omission and value faults.
          In [6], N-Version Programming is used to maximize the independence of version development and to employ
      design diversity in order to minimize the probability that two or more member versions will produce similar
      erroneous results. N-version programming is introduced two major fault tolerant software models; N-Version
      Software (NVS) and Recovery Block (RB) models. Next, the process of building N-version software is discussed in
      detail, including the specification, programming and execution support of NVS units.
          An N-version software (NVS) unit is a fault tolerant software unit that depends on a generic decision algorithm
      to determine a consensus result of two or more member versions of the NVS unit. The RB model applies the
      acceptance test to the output of an individual alternate service; this acceptance test must by necessity be specific for
      every distinct failure service.

                    III.          PROPOSED SYSTEM
          This paper describes one of the fault tolerant technique, i.e. Recovery Block model which can be used to provide
      fault tolerant service. An online monitoring approach is used to monitor the process as well as to detect the faults.
      Compared with the existing online monitoring efforts, our approach has the specific features.


 Department of CSE, Sun College of Engineering and Technology
National Conference on Role of Cloud Computing Environment in Green Communication 2012                                                                 408




                     Fig.1. Overall Architecture
          A monitoring model System event related to Web service behaviour is usually difficult to monitor because they
      are usually scattered in the whole system. This paper deals with this problem by introducing a monitoring model,
      systematically covering different kinds of events. A flexible monitoring mechanism, based on the monitoring model,
      that mechanism is able to generate automatically monitoring code from the constraint specification. The monitoring
      code can be deployed into the target system by administrator.
          It is capable of extracting relevant information about a Web service and its operating environment, and providing
      support for checking the service behaviour against its requirements. The monitoring framework is able to use the
      formally specified constraints to generate automatically monitoring code.
          The management centre providing as the improvement that they are constructing the secondary services, i.e.
      fault tolerant services. This service holds the copy of the application database. In case if there is delay in response or
      causes server down in application database means request message from the secondary service and retrieving the
      data very fast. It can be done by using RB model.
          The RB approach attempts to prevent the residual software service faults from impacting on the system
      environment, and it is aimed at providing fault-tolerant service which may be nested within a sequential program.
      The RB model applies the acceptance test to the output of an individual alternate software service. If the acceptance
      test is passed then the outcome of the service is regarded as successful and the recovery block can be exited. If the
      test fails or if any errors are detected by other means during the execution of the alternate service, then an exception
      is raised and backward error recovery is invoked to continue the operation. This restores the state of the system to
      what it was on entry. After such recovery, the next alternate service is executed and then the acceptance test is
      applied again. This sequence continues until either an acceptance test is passed by the user.

                  IV.           CONCLUSION
          The monitoring framework and RB model is applicable for providing the fault tolerant service even under the
      failure of some services. Here the overall architecture showed that, monitoring the web service by using the
      management centre and providing the fault tolerant service using RB model.
          The performance costs are acceptable considering the gains in the availability and reliability afforded by the RB
      model. Aiming future work, there is the implementation of other fault tolerant techniques and integration of the
      monitoring framework.

                      V.               REFRENCES

      [1] C.S. Langdon, “The State of Web Services,” Computer, vol. 36, no. 7, pp. 93-94, July 2003.

      [2] S. Degwekar, S.Y.W. Su, and H. Lam, “Constraint Specification and Processing in Web Services Publication and Discovery,” Proc. Second
      IEEE Int’l Conf. Web Services (ICWS ’04), pp. 210-217, 2004.

      [3] AW. Qianxiang, et al., “An online monitoring approach for Web service requirements,” IEEE Transactions on Services Computing, vol. 2,
      no. Copyright 2010, The Institute of Engineering and Technology, 2009, pp. 338-351
      [4] ] Dialini, V., Miles, S., Moreau, L. De Roure, D., Luck, M. (2002). Transparent Fault Tolerant for Web Services Based Architectures. Euro-
      Par 2002. Parallel Processing:8th InternationalEuro-Par Conference Paderborn, Germany Proceedings. Volume 2400

      [5] Santos GT, Lung LC, Montez C (2005) Ftweb: A fault tolerant infrastructure for web services. In: EDOC, pp 95{105}


 Department of CSE, Sun College of Engineering and Technology
National Conference on Role of Cloud Computing Environment in Green Communication 2012                                                       409



      [6] A.A. Avizienis, “The Methodology of N-version Programming”, Software Fault Tolerance, edited by M. Lyu, John Wiley & Sons, 1995.




 Department of CSE, Sun College of Engineering and Technology

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:7
posted:7/26/2012
language:
pages:4