Ramamurthy B., Upadhyaya S. J., Iyer R. K. An object-oriented test-bed for the evaluation of check-pointing and recovery systems. Proceedings of the 27th IEEE Int. Symposium on Fault-Tolerant Computing, Seattle, USA, 1997; 194-203.  Home/Search   Document Details and Download   Summary   Related Articles   Check

.... styles for implementing application transparent rollback recovery in messagepassing systems, namely coordinated checkpointing, message logging, and communicationinduced checkpointing (CIC) 5] Both coordinated checkpointing and message logging have received considerable analysis in the literature [4,6,11,14,17,18,20,21,24,28], but little is known about the behavior of CIC protocols. This paper presents an experimental analysis of these protocols through a prototype implementation and reveals several of their theoretical and pragmatic characteristics. CIC protocols are not new. The paper by Briatico et al. was perhaps ....

B. Ramamurthy and S.J. Upadhyaya and R.K. Iyer. An object-oriented testbed for the evaluation of checkpointing and recovery Systems. In Proceedings of the 27th IEEE Fault-Tolerant Computing Symposium, pp. 194---203, Jun. 1997.

....[19] MIST [5] Co check [31] Manetho [9] and libft [13] As opposed to Egida , however, these systems are designed to support only a specific checkpointing or message logging protocol. Egida emphasis on code reusability and extensibility is also present in the recently introduced OTEC simulator [24]. OTEC s object oriented design provides some degree of code reusability and can be used to compare the performance of different checkpointing and recovery protocols, but only through simulations. The run time system for rollback recovery that is closest to Egida in terms of extensibility is RENEW ....

B. Ramamurthy, S. J. Upadhyaya, and R. K. Iyer. An Object-Oriented Testbed for the Evaluation of Checkpointing and Recovery Systems. In Proceedings of the 27th IEEE Fault-Tolerant Computing Symposium, pages 194--203, June 1997.

....specific SWFT mechanisms, and multithreading enables solutions to a number of difficult problems in a distributed environment. A research area related to Juleps focus is the construction of simulated environments enabled to analyze distributed process recovery protocols. One such project is OTEC [9], an object oriented testbed, which utilizes the DEPEND [6] system simulation tool with fault injection capability. The distributed application to be tested with the OTEC system could be a version simulated from user specifications, or the actual implemented application. OTEC provides a set of ....

B. Ramamurthy, S. J. Upadhyaya, and R. K. Iyer, An Object-Oriented Testbed for the Evaluation of Checkpointing and Recovery Systems. In 27 th International Symposium on FaultTolerant Computing (FTCS 27), pages 194-203, June, 1997.

....requirements of their application. Finally, Egida enables designers of fault tolerance protocols to develop new rollback recovery protocols by combining different implementations of the core functionalities in novel ways. Egida shares some of the design goals of other frameworks such as OTEC [20] and RENEW [18] for implementing rollback recovery protocols. Egida is unique in that it allows rollback recovery protocols to be configured from basic building blocks. This approach is similar to the one used in other systems for instance, in Adaptive [23] and xkernel [13] for networking ....

....[16] MIST [5] Co check [26] Manetho [9] and libft [12] As opposed to Egida, however, these systems are designed to support only a specific checkpointing or message logging protocol. Egida s emphasis on code reusability and extensibility is also present in the recently introduced OTEC simulator [20]. OTEC s object oriented design provides some degree of code reusability and can be used to compare the performance of different checkpointing protocols, but only through simulations (for example, varying message rates and checkpoint sizes) The run time system for rollback recovery that is ....

B. Ramamurthy, S. J. Upadhyaya, and R. K. Iyer. An Object-Oriented Testbed for the Evaluation of Checkpointing and Recovery Systems. In Proceedings of the 27th IEEE Fault-Tolerant Computing Symposium, pages 194--203, June 1997.

....appears more Poisson ish than the other networks, which is reflected not only in its TTF distribution pattern, but also in the closer matching of simulated values to the equations of Section 3. However, it is a more interesting issue to provide a rigorous statistical analysis to the data. Like [22], we have employed a simulator to draw conclusions that are only as good as the simulator s base assumptions. Thus, questions also may be raised concerning the validity of the simulations. For example, all uptimes are considered equivalent by the simulator, whereas in real life, machines undergo ....

B. Ramamurthy, S. J. Upadhyaya, and R. K. Iyer. An objectoriented testbed for the evaluation of checkpointing and recovery systems. In 27th Int. Symp. on Fault-Tolerant Comp., 1997.

.... protocols for distributed systems can be divided into a small number of classes [1, 10] uncoordinated [11] coordinated [12 15] communication induced [4, 16] and message logging [5, 17, 18] OTEC is an object oriented simulator designed for the analysis of checkpointing and recovery techniques [19]. It uses DEPEND [20] and SIMPAR [21] for dependability and reliability evaluation in multicomputer systems. OTEC has a set of predefined classes for checkpointing, error detection and recovery, which can then be reused and composed to build new checkpoint protocols. The evaluation of the ....

B. Ramamurthy, S. Upadhyaya, and R. Iyer, "An objectoriented testbed for the evaluation of checkpointing and recovery systems", Proc. of the Inter. Symp. on Fault-Tolerant Comp., pp. 194--203, Jun. 1997.

....along with general purpose simulation support. We found such a tool in DEPEND [20] which is a dependability analysis tool based on the robust simulation package CSIM [21] With DEPEND as the core, we built an object oriented testbed for simulation of checkpointing and recovery called OTEC [22] to allow for design, development and comparison of different checkpointing and recovery schemes in a common test environment under realistic fault scenarios. OTEC reuses and extends many of the facilities provided by DEPEND and CSIM. OTEC is being used for the simulation reported here. 6.1 ....

B. Ramamurthy, S. Upadhyaya, and R. K. Iyer, "An object-oriented testbed for the evaluation of checkpointing and reco very systems," Dig. Papers 27th Annual IEEE International Symposium on Fault Tolerant Com puting (FTCS), pp. 194--203, June 1997.

No context found.

Ramamurthy B., Upadhyaya S. J., Iyer R. K. An object-oriented test-bed for the evaluation of check-pointing and recovery systems. Proceedings of the 27th IEEE Int. Symposium on Fault-Tolerant Computing, Seattle, USA, 1997; 194-203.

No context found.

B. Ramamurthy, S. J. Upadhyaya, R. K. Iyer, An object-oriented test-bed for the evaluation of checkpointing and recovery systems, Proceedings of the 27th International Symposium on Fault-Tolerant Computing, IEEE, Seattle, WA, USA, 1997, pp. 194-203

No context found.

B. Ramamurthy, S. J. Upadhyaya and R. K. Iyer, "An object-oriented test-bed for the evaluation of check-pointing and recovery systems", Proceedings of the 27th International Symposium on Fault-Tolerant Computing, IEEE, Seattle, WA, USA, 1997, pp. 194-203.

No context found.

B. Ramamurthy, S. Upadhyaya and R.K. Iyer. "An object-oriented testbed for the evaluation of checkpointing and recovery systems." In Proceedings of the Twenty Seventh International Symposium on Fault-Tolerant Computing (FTCS-27), pp. 194---203, Jun. 1997.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC