Carlos Guilherme Crisóstomo Bartolomeu

Dynamic Adaptation of Geo-Replicated CRDTs

Tese submetida para provas de mestrado em Engenharia Informática e de Computadores Instituto Superior Técnico, Universidade de Lisboa.

Abstract

With the advent of cloud computing, and the need to maintain data replicated in geographically remote data centers, searching for strategies to provide data consistency with minimal synchronization became very relevant. Unfortunately, most data types require operations to be totally ordered to ensure replica consistency.

Conflict-free Replicated Data Types (CRDTs) are data types whose operations do not conflict with each other and, therefore, can be replicated with minimal coordination among replicas. While it is easy to ensure that all replicas of CRDTs become eventually consistent when the system becomes quiescent, different techniques can be used to propagate the updates as efficiently as possible. Different approaches, such as state transfer and operation forwarding, have been proposed to propagate the updates as efficiently as possible, with different tradeoffs among the amount of network traffic generated and the staleness of local information.

This thesis proposes and evaluates techniques to automatically adapt a CRDT implementation, such that the best approach is used, based on the application needs (captured by a SLA) and the observed system configuration. Our techniques have been integrated in SwiftCloud, a state of the art geo-replicated store based on CRDTs.

Publicações

Dynamic Adaptation of Geo-Replicated CRDTs: Carlos Guilherme Crisóstomo Bartolomeu; MSc Thesis. Instituto Superior Técnico, Universidade de Lisboa.; October, 2015.; Available BibTeX, MSC Thesis, extended abstract of the thesis, and mid-term report.
Dynamic Adaptation of Geo-Replicated CRDTs.: C. Bartolomeu, M. Bravo, L. Rodrigues.; In Proceedings of the 31st ACM/SIGAPP Symposium on Applied Computing (SAC), Dependable and Adaptive Distributed Systems Track, Pisa, Italy April 4-8, 2016.

Luís Rodrigues