Revisiting Certification-Based Replicated Database Recovery

Ruiz-Fuertes, M. I.; Pla-Civera, J.; Armendáriz-Iñigo, José Enrique; Mendívil, José Ramón González de; Muñoz-Escoí, Francesc D.

doi:10.1007/978-3-540-76848-7_35

Cited by 5 publications

(9 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We further enhanced it by adding two other previously proposed techniques meant to improve performance. Namely, the ability to do parallel recovery [10] by having more than one replica contributing with its state and the introduction of convergence phases [2,23]. Our complete algorithm still comprehends basic optimizations such as purging redundant data changes and the compression of the transmitted data.…”

Section: Recovery Protocolmentioning

confidence: 99%

“…The data transfer is done with a unicast channel and using convergence phases. It has been shown in [23] that recovery time can be substantially reduced if split in convergence phases. During the whole recovery process, the recovering replica does not handle any user requests nor it participates in the replication protocol.…”

Section: Recovery Protocolmentioning

confidence: 99%

“…1 this process can be generalized to n phases. This has been shown [23] to outperform the alternative of having the recovering replica buffering the delivered write sets until the missing log is applied. This buffer would then go through the certification process and, when due, the updates applied.…”

Section: Recovery Protocolmentioning

confidence: 99%

“…All machines ran Linux (with kernel version 2.6.22-14-smp). This hardware configuration corresponds to commodity servers and was chosen to be on par with the systems used in recent related work [17,22,18,23,16].…”

Section: Experimental Settingmentioning

confidence: 99%

“…Recently, a large body of research has been dedicated to the online recovery of replicas in database clusters [13,2,9,10,23,16]. These works are all based on a reliable group communication substrate and have in common the use of consistent database replication protocols.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

On the Cost of Database Clusters Reconfiguration

Vilaça

Pereira

Oliveira

et al. 2009

2009 28th IEEE International Symposium on Reliable Distributed Systems

View full text Add to dashboard Cite

Database clusters based on share-nothing replication techniques are currently widely accepted as a practical solution to scalability and availability of the data tier. A key issue when planning such systems is the ability to meet service level agreements when load spikes occur or cluster nodes fail. This translates into the ability to provision and deploy additional nodes. Many current research efforts focus on designing autonomic controllers to perform such reconfiguration, tuned to quickly react to system changes and spawn new replicas based on resource usage and performance measurements. In contrast, we are concerned about the inherent impact of deploying an additional node to an online cluster, considering both the time required to finish such an action as well as the impact on resource usage and performance of the cluster as a whole. If noticeable, such impact hinders the practicability of self-management techniques, since it adds an additional dimension that has to be accounted for. Our approach is to systematically benchmark a number of different reconfiguration scenarios to assess the cost of bringing a new replica online. We consider factors such as: workload characteristics, incremental and parallel recovery, flow control and outdatedness of the recovering replica. As a result, we show that research should be refocused from optimizing the capture and transmition of changes to applying them, which in a realistic setting dominates the cost of the recovery operation.

show abstract

Section: Recovery Protocolmentioning

confidence: 99%

Section: Recovery Protocolmentioning

confidence: 99%

Section: Recovery Protocolmentioning

confidence: 99%