High performance state-machine replication

10 years 6 months ago
High performance state-machine replication
Computer systems are usually made fault tolerant through replication. By replicating a service on multiple servers we make sure that if some replicas fail, the service can still be performed by the operational nodes. However once a service is replicated, ensuring consistency among all the copies is not easy. Conflicting requests submitted simultaneously to different nodes can lead to inconsistent states in the replicas and therefore meaningless replies to clients. My research addresses the design and implementation of one of the pillars of replication, atomic broadcast. The goal is to propose solutions for high performance atomic broadcast algorithms that are capable of addressing consistency with low latency and high throughput. Research Advisor Prof. Fernando Pedone Academic Advisor Review Committee Prof. Fernando Pedone Prof. Cesare Pautasso, Prof. Laura Pozzi Research Advisor’s approval (Prof. Fernando Pedone): Date: ............................... PhD Director’s approval (Pr...
Parisa Jalili Marandi, Marco Primi, Fernando Pedon
Added 19 Dec 2011
Updated 19 Dec 2011
Type Journal
Year 2011
Where DSN
Authors Parisa Jalili Marandi, Marco Primi, Fernando Pedone
Comments (0)