Publicação

Probabilistic data types

Ver documento

Detalhes bibliográficos
Resumo:Conflict-Free Replicated Data Types (CRDTs) provide deterministic outcomes from concurrent executions. The conflict resolution mechanism uses information on the ordering of the last operations performed, which indicates if a given operation is known by a replica, typically using some variant of version vectors. This thesis will explore the construction of CRDTs that use a novel stochastic mechanism that can track with high accuracy knowledge of the occurrence of recently performed operations and with less accuracy for older operations. The aim is to obtain better scaling properties and avoid the use of metadata that is linear on the number of replicas.
Autores principais:Fernandes, Pedro Henrique Moreira Gomes
Assunto:Conflict-Free Replicated Data Types Probabilistic representation of sets Bloom filters Eventual consistency Representação probabilística de conjuntos Consistencia eventual
Ano:2021
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso aberto
Instituição associada:Universidade do Minho
Idioma:inglês
Origem:RepositóriUM - Universidade do Minho
Descrição
Resumo:Conflict-Free Replicated Data Types (CRDTs) provide deterministic outcomes from concurrent executions. The conflict resolution mechanism uses information on the ordering of the last operations performed, which indicates if a given operation is known by a replica, typically using some variant of version vectors. This thesis will explore the construction of CRDTs that use a novel stochastic mechanism that can track with high accuracy knowledge of the occurrence of recently performed operations and with less accuracy for older operations. The aim is to obtain better scaling properties and avoid the use of metadata that is linear on the number of replicas.