Cookies
Usamos cookies para melhorar nosso site e a sua experiência. Ao continuar a navegar no site, você aceita a nossa política de cookies. Ver mais
Fechar
  • Menu
Sobre

Sobre

Em termos de temas de investigação principais destaca-se a gestão de dados em modelos de coerência fraca, mecanismos de agregação de dados e causalidade em sistemas distribuídos. No últimos anos, e em colaboração outros investigadores, têm sido desenvolvidos mecanismos de sumarização de dados como os Scalable Bloom Filters, registo de causalidade em ambientes dinâmicos com Interval Tree Clocks e Dotted Version Vectors, bem como abordagens robustas para o suporte à alta disponibilidade com coerência fraca via Conflict-Free Replicated Data Types. Alguns destes mecanismos têm sido aplicados na base de dados distribuída Riak e no Akka distributed data, estando estes mesmos em uso em diversas aplicações finais com milhões de utilizadores a nível global.

Tópicos
de interesse
Detalhes

Detalhes

  • Nome

    Carlos Baquero
  • Cluster

    Informática
  • Cargo

    Responsável de Área
  • Desde

    01 novembro 2011
003
Publicações

2019

Scalable eventually consistent counters over unreliable networks

Autores
Almeida, PS; Baquero, C;

Publicação
Distributed Computing

Abstract

2019

Conflict-Free Replicated Data Types CRDTs

Autores
Preguiça, NM; Baquero, C; Shapiro, M;

Publicação
Encyclopedia of Big Data Technologies.

Abstract

2019

Efficient synchronization of state-based CRDTs

Autores
Enes, V; Almeida, PS; Baquero, C; Leitao, J;

Publicação
Proceedings - International Conference on Data Engineering

Abstract
To ensure high availability in large scale distributed systems, Conflict-free Replicated Data Types (CRDTs) relax consistency by allowing immediate query and update operations at the local replica, with no need for remote synchronization. State-based CRDTs synchronize replicas by periodically sending their full state to other replicas, which can become extremely costly as the CRDT state grows. Delta-based CRDTs address this problem by producing small incremental states (deltas) to be used in synchronization instead of the full state. However, current synchronization algorithms for delta-based CRDTs induce redundant wasteful delta propagation, performing worse than expected, and surprisingly, no better than state-based. In this paper we: 1) identify two sources of inefficiency in current synchronization algorithms for delta-based CRDTs; 2) bring the concept of join decomposition to state-based CRDTs; 3) exploit join decompositions to obtain optimal deltas and 4) improve the efficiency of synchronization algorithms; and finally, 5) experimentally evaluate the improved algorithms. © 2019 IEEE.

2018

Delta State Replicated Data Types

Autores
Almeida, PS; Shoker, A; Baquero, C;

Publicação
JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING

Abstract

2018

Global-Local View: Scalable Consistency for Concurrent Data Types

Autores
Akkoorath, DD; Brandão, J; Bieniusa, A; Baquero, C;

Publicação
Euro-Par 2018: Parallel Processing - 24th International Conference on Parallel and Distributed Computing, Turin, Italy, August 27-31, 2018, Proceedings

Abstract
Concurrent linearizable access to shared objects can be prohibitively expensive in a high contention workload. Many applications apply ad-hoc techniques to eliminate the need for synchronous atomic updates, which may result in non-linearizable implementations. We propose a new model which leverages such patterns for concurrent access to objects in a shared memory system. In this model, each thread maintains different views on the shared object: a thread-local view and a global view. As the thread-local view is not shared, it can be updated without incurring synchronization costs. These local updates become visible to other threads only after the thread-local view is merged with the global view. This enables better performance at the expense of linearizability. We discuss the design of several datatypes and evaluate their performance and scalability compared to linearizable implementations. © 2018, Springer International Publishing AG, part of Springer Nature.

Teses
supervisionadas

2017

Broadcast Cancellation in Unstructured Networks

Autor
Rui Manuel Dias Ferreira Lima

Instituição
UM

2017

Efficient Synchronization of state-based CRDTs

Autor

Instituição
UM

2017

Sensing, coordination and actuation in office spaces

Autor

Instituição
UM

2016

Optimizing Operation-based Conflict-free Replicated Data Types

Autor
Georges Younes

Instituição
UM

2016

Broadcast Cancellation

Autor
Rui Manuel Dias Ferreira Lima

Instituição
UM