Crash recovery with partial amnesia failure model issues.

Marín, Rubén De Juan

doi:10.4995/thesis/10251/3302

Cited by 3 publications

(2 citation statements)

References 83 publications

(204 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Partial recovery is possible thanks to virtual synchrony [8]; however, this property provided by the GCS (which ensures that processes that participate in the same two consecutive views deliver the same set of multicast messages in the former view) expresses delivery guarantees that have nothing to do with processing. Consequently, the real state at the joining replica may differ from the last state it is assumed it had before crashing, due to the fact that it may not have processed all delivered messages, causing the amnesia phenomenon [10,19]. Thus, the joining replica will have to obtain two types of lost messages: forgotten messages that were delivered but not applied before failure, and missed messages that were delivered at the system during the disconnection period.…”

Section: Introductionmentioning

confidence: 86%

See 1 more Smart Citation

An implementation of a replicated file server supporting the crash-recovery failure model

et al. 2010

View full text Add to dashboard Cite

Data replication techniques are widely used for improving availability in software applications. Replicated systems have traditionally assumed the fail-stop model, which limits fault tolerance. For this reason, there is a strong motivation to adopt the crash-recovery model, in which replicas can dynamically leave and join the system. With the aim to point out some key issues that must be considered when dealing with replication and recovery, we have implemented a replicated file server that satisfies the crash-recovery model, making use of a Group Communication System. According to our experiments, the most interesting results are that the type of replication and the number of replicas must be carefully determined, specially in update intensive scenarios; and, the variable overhead imposed by the recovery protocol to the system. From the latter, it would be convenient to adjust the desired trade-off between recovery time and system throughput in terms of the service state size and the number of missed operations.

show abstract

Section: Introductionmentioning

confidence: 86%

“…The recovery of outdated replicas can be carried out in many ways ranging from total [3] to partial recovery [10,[18][19][20]32]. Let us see their main differences with a concrete example considering a replicated file server where a replica joins in during normal operation.…”

Section: Introductionmentioning

confidence: 99%

An implementation of a replicated file server supporting the crash-recovery failure model

et al. 2010

View full text Add to dashboard Cite

show abstract

Optimizing recovery protocols for replicated database systems

Muñoz¹,

Hector²

View full text Add to dashboard Cite

AgradecimientosEn primer lugar quiero agradecer a quien considero un gran profesor y amigo, mi director de tesis Dr. Francisco Muñoz, gracias Paco por todo tu apoyo y paciencia durante el desarrollo de este trabajo, gracias por mostrarme lo que la dedicación y empeño pueden lograr, gracias por todo el conocimiento que has compartido conmigo y tu gran capacidad para tener siempre una pronta y acertada respuesta.Gracias a Enrique Armendáriz un gran compañero y también director de tesis. Gracias Enrique por esa gran capacidad para generar ideas y llevarlas a la práctica y por tu orientación en este interesante y algunas vecesárido camino de la investigación.Ambos son un gran ejemplo a seguir.Gracias al Instituto Tecnológico de Informática por haberme permitido colaborar en el grupo de investigación de Sistemas Distribuidos. Gracias a todos los miembros del grupo, en particular quiero dar las gracias a Rubén de Juan Marín, Idoia Ruiz, Jerónimo Pla y Hendrik Decker con quienes he colaborado para la realización de los trabajos de investigación.Gracias también a Pedro Blesa cuyas primeras instrucciones en el campo de la investigación fueron fundamentales.Gracias a mis compañeros del Instituto Tecnológico de la Laguna, en especial a Roberto Dominguez por su apoyo invaluable.En el apartado personal quiero dar las gracias a toda mi familia, a mis padres Luis y Lidia, a quienes debo todo lo que soy. A mis hermanos que siempre han creído en mí. A mis suegros José y Consuelo en quienes he encontrado unos segundos padres.A Mayela, mi amada esposa, no encuentro palabras para agradecerte todo el apoyo que me has dado, siempre conmigo y alentándome cuando el camino parece indescifrable. A mis hijos por su cariño y comprensión. A Paty y Victor por su apoyo incondicional.

show abstract