{"title":"On Markov reward modelling with FSPNs","authors":"K. Wolter, A. Zisowsky","doi":"10.1109/IPDS.2000.839459","DOIUrl":"https://doi.org/10.1109/IPDS.2000.839459","url":null,"abstract":"In this paper fluid stochastic Petri nets (FSPNs) will be used for modelling reward in a performability model. Two variations of a known performability model are presented in order to demonstrate the ability of FSPNs in modelling accumulated rate reward as well as accumulated impulse reward. In the first model two fluid places are used, one of which represents the profit (reward) obtained by operating the system and the other one the buffet; that is approximated continuously. In the second model only one fluid place is used, representing the costs (negative reward) arising due to repair of system components. The costs increase continuously at deterministic rate while the system is in state of repair (which is a rate reward in the model). Additional costs incur each time the buffer fails (which is an impulse reward in the model). With a numerical solution algorithm the distribution of the reward and its mean are computed.","PeriodicalId":162523,"journal":{"name":"Proceedings IEEE International Computer Performance and Dependability Symposium. IPDS 2000","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122719828","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Hybrid analysis of SGSPNs with time-dependent transition rates","authors":"P. Buchholz","doi":"10.1109/IPDS.2000.839476","DOIUrl":"https://doi.org/10.1109/IPDS.2000.839476","url":null,"abstract":"For the class Superposed Generalized Stochastic Petri Nets (SGSPNs) we present a new analysis algorithm which combines the ideas of transient numerical analysis via uniformization and discrete event simulation. Simulation is used to determine the firing times of synchronized transitions describing interactions between the components of a SGSPN. Numerical analysis is used to compute the distribution of states between firing times of synchronized transitions. The approach avoids handling vectors of the size of the state space and can therefore be used for models with a huge state space which cannot be analyzed with numerical analysis techniques. On the other hand, the approach allows the computation of reliable estimates for results based on small probabilities or rare events which cannot be analyzed with simulation. As a by product of the handling of synchronized transitions using simulation, it is possible to extend the class of SGSPNs by allowing synchronized transitions with time dependent transition rates.","PeriodicalId":162523,"journal":{"name":"Proceedings IEEE International Computer Performance and Dependability Symposium. IPDS 2000","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125441959","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"I/O phase characterization of TPC-H query operations","authors":"M. Kandaswamy, R. L. Knighten","doi":"10.1109/IPDS.2000.839466","DOIUrl":"https://doi.org/10.1109/IPDS.2000.839466","url":null,"abstract":"With the introduction of newer faster processors, speedier interconnects and sophisticated I/O system designs, the processing capabilities of servers built in current market place are ever increasing, but so are the myriad commercial applications that they are used for. Unfortunately, detailed studies capturing the demands of commercial applications/workloads have been very few. In this paper we characterize the I/O activity of the 22 I/O-intensive database queries of the commercial decision support system (DSS) benchmark, namely, TPC-H and a smaller set of queries called \"DSS-lite\" that perform individual database operations. TPC-H query behavior varies throughout its execution but we identify \"phases\" in a query during which the performance is close forming a steady state. We use the DSS-lite results and query plan information to decipher the various phases of the TPC-H queries and characterize their I/O requirements, which can aid in forecasting or planning a DSS system's I/O capacity.","PeriodicalId":162523,"journal":{"name":"Proceedings IEEE International Computer Performance and Dependability Symposium. IPDS 2000","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128011016","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"On the effectiveness of a message-driven confidence-driven protocol for guarded software upgrading","authors":"A. Tai, K. Tso, L. Alkalai, S. Chau, W. Sanders","doi":"10.1109/IPDS.2000.839464","DOIUrl":"https://doi.org/10.1109/IPDS.2000.839464","url":null,"abstract":"In order to accomplish dependable onboard evolution, we develop a methodology which is called \"guarded software upgrading\" (GSU). The core of the methodology is a low-cost error containment and recovery protocol that escorts an upgraded software component through onboard validation, and guarded operation, safeguarding mission functions. The message-driven confidence-driven (MDCD) nature of the protocol eliminates the need for costly process coordination or atomic action, yet guarantees that the system will reach a consistent global state upon the completion of the rollback or roll-forward actions carried out by individual processes during error recovery. To validate the ability of the MDCD protocol to enhance system reliability when a software component undergoes onboard upgrading in a realistic, non-ideal environment, we conduct a stochastic activity network model based analysis. The results confirm the effectiveness of the protocol as originally surmised. Moreover, the model-based analyse's provides useful insight about the system behavior resulting from the use of the protocol under various conditions in its execution environment, facilitating effective use of the protocol.","PeriodicalId":162523,"journal":{"name":"Proceedings IEEE International Computer Performance and Dependability Symposium. IPDS 2000","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134438316","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. Stott, Benjamin Floering, Daniel Burke, Z. Kalbarczyk, Ravishankar K. Iyer
{"title":"NFTAPE: a framework for assessing dependability in distributed systems with lightweight fault injectors","authors":"D. Stott, Benjamin Floering, Daniel Burke, Z. Kalbarczyk, Ravishankar K. Iyer","doi":"10.1109/IPDS.2000.839467","DOIUrl":"https://doi.org/10.1109/IPDS.2000.839467","url":null,"abstract":"Many fault injection tools are available for dependability assessment. Although these tools are good at injecting a single fault model into a single system, they suffer from two main limitations for use in distributed systems: (1) no single tool is sufficient for injecting all necessary fault models; (2) it is difficult to port these tools to new systems. NFTAPE, a tool for composing automated fault injection experiments from available lightweight fault injectors, triggers, monitors, and other components, helps to solve these problems. We have conducted experiments using NFTAPE with several types of lightweight fault injectors, including driver-based, debugger-based, target-specific, simulation-based, hardware-based, and performance-fault injections. Two example experiments are described in this paper. The first uses a hardware fault injector with a Myrinet LAN; the other uses a Software Implemented Fault Injection (SWIFI) fault injector to target a space-imaging application.","PeriodicalId":162523,"journal":{"name":"Proceedings IEEE International Computer Performance and Dependability Symposium. IPDS 2000","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128103406","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"On building non-intrusive performance instrumentation blocks for CORBA-based distributed systems","authors":"B. Sridharan, B. Dasarathy, A. Mathur","doi":"10.1109/IPDS.2000.839472","DOIUrl":"https://doi.org/10.1109/IPDS.2000.839472","url":null,"abstract":"A non-intrusive, reusable framework for collecting performance statistics of CORBA-based distributed systems is proposed. The usefulness of the framework in the context of comparing performance improvement techniques is described. Also discussed is application of this framework to a large telecommunications application and experimental results on its relative overhead.","PeriodicalId":162523,"journal":{"name":"Proceedings IEEE International Computer Performance and Dependability Symposium. IPDS 2000","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130513154","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Mission time analysis of large dependable systems","authors":"D. Rácz, M. Telek","doi":"10.1109/IPDS.2000.839460","DOIUrl":"https://doi.org/10.1109/IPDS.2000.839460","url":null,"abstract":"The mission time analysis of large dependable systems that can be described by Markov Reward Models (MRM) with phase type distributed impulse and constant rate rewards is considered in the paper. A single Laplace transform domain description of the distribution of completion time is provided through a new analysis approach. Based on this description an effective numerical method is introduced which allows the evaluation of models with large state space (/spl sim/10/sup 6/ states). The applied analysis approach makes the use of an expanded Markov chain, but the state space expansion is much less than for common \"phase type expansion\", because the expanded state space is composed by the union (instead of product) of the original state space and the state space of the phase type structure of non-zero impulse rewards. (Roughly speaking, the applied state space expansion is additive in contrast with the multiplicative state space expansion used for phase type modeling.) The proposed method, which is a counterpart of the analysis method of accumulated reward of MRMs with rate and impulse rewards, provides the moments of reward measures approximately on the same computational cost and memory requirement as the transient analysis of the expanded Continuous Time Markov Chain. Numerical example demonstrates the abilities of the proposed method.","PeriodicalId":162523,"journal":{"name":"Proceedings IEEE International Computer Performance and Dependability Symposium. IPDS 2000","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129264608","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}