arXiv subjects

Carlos Baquero

Publications and source records attributed to Carlos Baquero.

At least 37 records · Page 2Linked to original sources

arXiv2019

Efficient Synchronization of State-based CRDTs

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 synchronisation 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) evaluate the improved algorithms.

cs.DC↗

arXiv2018

Conflict-free Replicated Data Types (CRDTs)

A conflict-free replicated data type (CRDT) is an abstract data type, with a well defined interface, designed to be replicated at multiple processes and exhibiting the following properties: (1) any replica can be modified without coordinating with another replicas; (2) when any two replicas have received the same set of updates, they reach the same state, deterministically, by adopting mathematically sound rules to guarantee state convergence.

cs.DC↗

arXiv2017

Pure Operation-Based Replicated Data Types

Distributed systems designed to serve clients across the world often make use of geo-replication to attain low latency and high availability. Conflict-free Replicated Data Types (CRDTs) allow the design of predictable multi-master replication and support eventual consistency of replicas that are allowed to transiently diverge. CRDTs come in two flavors: state-based, where a state is changed locally and shipped and merged into other replicas; operation-based, where operations are issued locally and reliably causal broadcast to all other replicas. However, the standard definition of op-based CRDTs is very encompassing, allowing even sending the full-state, and thus imposing storage and dissemination overheads as well as blurring the distinction from state-based CRDTs. We introduce pure op-based CRDTs, that can only send operations to other replicas, drawing a clear distinction from state-based ones. Data types with commutative operations can be trivially implemented as pure op-based CRDTs using standard reliable causal delivery; whereas data types having non-commutative operations are implemented using a PO-Log, a partially ordered log of operations, and making use of an extended API, i.e., a Tagged Causal Stable Broadcast (TCSB), that provides extra causality information upon delivery and later informs when delivered messages become causally stable, allowing further PO-Log compaction. The framework is illustrated by a catalog of pure op-based specifications for classic CRDTs, including counters, multi-value registers, add-wins and remove-wins sets.

cs.DC↗

arXiv2017

Practical Evaluation of the Lasp Programming Model at Large Scale - An Experience Report

Programming models for building large-scale distributed applications assist the developer in reasoning about consistency and distribution. However, many of the programming models for weak consistency, which promise the largest scalability gains, have little in the way of evaluation to demonstrate the promised scalability. We present an experience report on the implementation and large-scale evaluation of one of these models, Lasp, originally presented at PPDP `15, which provides a declarative, functional programming style for distributed applications. We demonstrate the scalability of Lasp's prototype runtime implementation up to 1024 nodes in the Amazon cloud computing environment. It achieves high scalability by uniquely combining hybrid gossip with a programming model based on convergent computation. We report on the engineering challenges of this implementation and its evaluation, specifically related to operating research prototypes in a production cloud environment.

cs.DC↗

arXiv2017

Global-Local View: Scalable Consistency for Concurrent Data Types

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 of synchronous atomic updates, which may result in non-linearizable implementations. We propose a new programming 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 show that it is possible to maintain thread-local views and to perform merge efficiently for several data types and evaluate their performance and scalability compared to linearizable implementations. Further, we discuss the consistency semantics of the data types and the associated programming model.

cs.DC↗

arXiv2016

Worlds of Events: Deduction with Partial Knowledge about Causality

Interactions between internet users are mediated by their devices and the common support infrastructure in data centres. Keeping track of causality amongst actions that take place in this distributed system is key to provide a seamless interaction where effects follow causes. Tracking causality in large scale interactions is difficult due to the cost of keeping large quantities of metadata; even more challenging when dealing with resource-limited devices. In this paper, we focus on keeping partial knowledge on causality and address deduction from that knowledge. We provide the first proof-theoretic causality modelling for distributed partial knowledge. We prove computability and consistency results. We also prove that the partial knowledge gives rise to a weaker model than classical causality. We provide rules for offline deduction about causality and refute some related folklore. We define two notions of forward and backward bisimilarity between devices, using which we prove two important results. Namely, no matter the order of addition/removal, two devices deduce similarly about causality so long as: (1) the same causal information is fed to both. (2) they start bisimilar and erase the same causal information. Thanks to our establishment of forward and backward bisimilarity, respectively, proofs of the latter two results work by simple induction on length.

cs.DC↗

arXiv2016

Delta State Replicated Data Types

CRDTs are distributed data types that make eventual consistency of a distributed object possible and non ad-hoc. Specifically, state-based CRDTs ensure convergence through disseminating the entire state, that may be large, and merging it to other replicas; whereas operation-based CRDTs disseminate operations (i.e., small states) assuming an exactly-once reliable dissemination layer. We introduce Delta State Conflict-Free Replicated Data Types ($δ$-CRDTs) that can achieve the best of both worlds: small messages with an incremental nature, as in operation-based CRDTs, disseminated over unreliable communication channels, as in traditional state-based CRDTs. This is achieved by defining delta mutators to return a delta-state, typically with a much smaller size than the full state, that to be joined with both local and remote states. We introduce the $δ$-CRDT framework, and we explain it through establishing a correspondence to current state-based CRDTs. In addition, we present an anti-entropy algorithm for eventual convergence, and another one that ensures causal consistency. Finally, we introduce several $δ$-CRDT specifications of both well-known replicated datatypes and novel datatypes, including a generic map composition.

cs.DC↗

arXiv2015

Eventually Consistent Register Revisited

In order to converge in the presence of concurrent updates, modern eventually consistent replication systems rely on causality information and operation semantics. It is relatively easy to use semantics of high-level operations on replicated data structures, such as sets, lists, etc. However, it is difficult to exploit semantics of operations on registers, which store opaque data. In existing register designs, concurrent writes are resolved either by the application, or by arbitrating them according to their timestamps. The former is complex and may require user intervention, whereas the latter causes arbitrary updates to be lost. In this work, we identify a register construction that generalizes existing ones by combining runtime causality ordering, to identify concurrent writes, with static data semantics, to resolve them. We propose a simple conflict resolution template based on an application-predefined order on the domain of values. It eliminates or reduces the number of conflicts that need to be resolved by the user or by an explicit application logic. We illustrate some variants of our approach with use cases, and how it generalizes existing designs.

cs.DC↗

arXiv2015

Efficient State-based CRDTs by Delta-Mutation

CRDTs are distributed data types that make eventual consistency of a distributed object possible and non ad-hoc. Specifically, state-based CRDTs ensure convergence through disseminating the en- tire state, that may be large, and merging it to other replicas; whereas operation-based CRDTs disseminate operations (i.e., small states) assuming an exactly-once reliable dissemination layer. We introduce Delta State Conflict-Free Replicated Datatypes (δ-CRDT) that can achieve the best of both worlds: small messages with an incremental nature, as in operation-based CRDTs, disseminated over unreliable communication channels, as in traditional state-based CRDTs. This is achieved by defining δ-mutators to return a delta-state, typically with a much smaller size than the full state, that is joined to both: local and remote states. We introduce the δ-CRDT framework, and we explain it through establishing a correspondence to current state-based CRDTs. In addition, we present an anti-entropy algorithm that ensures causal consistency, and we introduce two δ-CRDT specifications of well-known replicated datatypes.

cs.DC↗

arXiv2013

SwiftCloud: Fault-Tolerant Geo-Replication Integrated all the Way to the Client Machine

Client-side logic and storage are increasingly used in web and mobile applications to improve response time and availability. Current approaches tend to be ad-hoc and poorly integrated with the server-side logic. We present a principled approach to integrate client- and server-side storage. We support mergeable and strongly consistent transactions that target either client or server replicas and provide access to causally-consistent snapshots efficiently. In the presence of infrastructure faults, a client-assisted failover solution allows client execution to resume immediately and seamlessly access consistent snapshots without waiting. We implement this approach in SwiftCloud, the first transactional system to bring geo-replication all the way to the client machine. Example applications show that our programming model is useful across a range of application areas. Our experimental evaluation shows that SwiftCloud provides better fault tolerance and at the same time can improve both latency and throughput by up to an order of magnitude, compared to classical geo-replication techniques.

cs.DC↗

arXiv2013

Scalable Eventually Consistent Counters over Unreliable Networks

Counters are an important abstraction in distributed computing, and play a central role in large scale geo-replicated systems, counting events such as web page impressions or social network "likes". Classic distributed counters, strongly consistent, cannot be made both available and partition-tolerant, due to the CAP Theorem, being unsuitable to large scale scenarios. This paper defines Eventually Consistent Distributed Counters (ECDC) and presents an implementation of the concept, Handoff Counters, that is scalable and works over unreliable networks. By giving up the sequencer aspect of classic distributed counters, ECDC implementations can be made AP in the CAP design space, while retaining the essence of counting. Handoff Counters are the first CRDT (Conflict-free Replicated Data Type) based mechanism that overcomes the identity explosion problem in naive CRDTs, such as G-Counters (where state size is linear in the number of independent actors that ever incremented the counter), by managing identities towards avoiding global propagation and garbage collecting temporary entries. The approach used in Handoff Counters is not restricted to counters, being more generally applicable to other data types with associative and commutative operations.

cs.DC↗

arXiv2013

Genetic Algorithm with a Local Search Strategy for Discovering Communities in Complex Networks

In order to further improve the performance of current genetic algorithms aiming at discovering communities, a local search based genetic algorithm GALS is here proposed. The core of GALS is a local search based mutation technique. In order to overcome the drawbacks of traditional mutation methods, the paper develops the concept of marginal gene and then the local monotonicity of modularity function Q is deduced from each nodes local view. Based on these two elements, a new mutation method combined with a local search strategy is presented. GALS has been evaluated on both synthetic benchmarks and several real networks, and compared with some presently competing algorithms. Experimental results show that GALS is highly effective and efficient for discovering community structure.

cs.SI↗

arXiv2013

Markov random walk under constraint for discovering overlapping communities in complex networks

Detection of overlapping communities in complex networks has motivated recent research in the relevant fields. Aiming this problem, we propose a Markov dynamics based algorithm, called UEOC, which means, 'unfold and extract overlapping communities'. In UEOC, when identifying each natural community that overlaps, a Markov random walk method combined with a constraint strategy, which is based on the corresponding annealed network (degree conserving random network), is performed to unfold the community. Then, a cutoff criterion with the aid of a local community function, called conductance, which can be thought of as the ratio between the number of edges inside the community and those leaving it, is presented to extract this emerged community from the entire network. The UEOC algorithm depends on only one parameter whose value can be easily set, and it requires no prior knowledge on the hidden community structures. The proposed UEOC has been evaluated both on synthetic benchmarks and on some real-world networks, and was compared with a set of competing algorithms. Experimental result has shown that UEOC is highly effective and efficient for discovering overlapping communities.

cs.SI↗

arXiv2012

An optimized conflict-free replicated set

Eventual consistency of replicated data supports concurrent updates, reduces latency and improves fault tolerance, but forgoes strong consistency. Accordingly, several cloud computing platforms implement eventually-consistent data types. The set is a widespread and useful abstraction, and many replicated set designs have been proposed. We present a reasoning abstraction, permutation equivalence, that systematizes the characterization of the expected concurrency semantics of concurrent types. Under this framework we present one of the existing conflict-free replicated data types, Observed-Remove Set. Furthermore, in order to decrease the size of meta-data, we propose a new optimization to avoid tombstones. This approach that can be transposed to other data types, such as maps, graphs or sequences.

cs.DC↗

arXiv2012

Spectra: Robust Estimation of Distribution Functions in Networks

Distributed aggregation allows the derivation of a given global aggregate property from many individual local values in nodes of an interconnected network system. Simple aggregates such as minima/maxima, counts, sums and averages have been thoroughly studied in the past and are important tools for distributed algorithms and network coordination. Nonetheless, this kind of aggregates may not be comprehensive enough to characterize biased data distributions or when in presence of outliers, making the case for richer estimates of the values on the network. This work presents Spectra, a distributed algorithm for the estimation of distribution functions over large scale networks. The estimate is available at all nodes and the technique depicts important properties, namely: robust when exposed to high levels of message loss, fast convergence speed and fine precision in the estimate. It can also dynamically cope with changes of the sampled local property, not requiring algorithm restarts, and is highly resilient to node churn. The proposed approach is experimentally evaluated and contrasted to a competing state of the art distribution aggregation technique.

cs.DC↗

arXiv2011

Fast Distributed Computation of Distances in Networks

This paper presents a distributed algorithm to simultaneously compute the diameter, radius and node eccentricity in all nodes of a synchronous network. Such topological information may be useful as input to configure other algorithms. Previous approaches have been modular, progressing in sequential phases using building blocks such as BFS tree construction, thus incurring longer executions than strictly required. We present an algorithm that, by timely propagation of available estimations, achieves a faster convergence to the correct values. We show local criteria for detecting convergence in each node. The algorithm avoids the creation of BFS trees and simply manipulates sets of node ids and hop counts. For the worst scenario of variable start times, each node i with eccentricity ecc(i) can compute: the node eccentricity in diam(G)+ecc(i)+2 rounds; the diameter in 2*diam(G)+ecc(i)+2 rounds; and the radius in diam(G)+ecc(i)+2*radius(G) rounds.

cs.DC↗

arXiv2011

A Survey of Distributed Data Aggregation Algorithms

Distributed data aggregation is an important task, allowing the decentralized determination of meaningful global properties, that can then be used to direct the execution of other applications. The resulting values result from the distributed computation of functions like COUNT, SUM and AVERAGE. Some application examples can found to determine the network size, total storage capacity, average load, majorities and many others. In the last decade, many different approaches have been proposed, with different trade-offs in terms of accuracy, reliability, message and time complexity. Due to the considerable amount and variety of aggregation algorithms, it can be difficult and time consuming to determine which techniques will be more appropriate to use in specific settings, justifying the existence of a survey to aid in this task. This work reviews the state of the art on distributed data aggregation algorithms, providing three main contributions. First, it formally defines the concept of aggregation, characterizing the different types of aggregation functions. Second, it succinctly describes the main aggregation techniques, organizing them in a taxonomy. Finally, it provides some guidelines toward the selection and use of the most relevant techniques, summarizing their principal characteristics.

cs.DC↗

arXiv2011

Fault-Tolerant Aggregation: Flow-Updating Meets Mass-Distribution

Flow-Updating (FU) is a fault-tolerant technique that has proved to be efficient in practice for the distributed computation of aggregate functions in communication networks where individual processors do not have access to global information. Previous distributed aggregation protocols, based on repeated sharing of input values (or mass) among processors, sometimes called Mass-Distribution (MD) protocols, are not resilient to communication failures (or message loss) because such failures yield a loss of mass. In this paper, we present a protocol which we call Mass-Distribution with Flow-Updating (MDFU). We obtain MDFU by applying FU techniques to classic MD. We analyze the convergence time of MDFU showing that stochastic message loss produces low overhead. This is the first convergence proof of an FU-based algorithm. We evaluate MDFU experimentally, comparing it with previous MD and FU protocols, and verifying the behavior predicted by the analysis. Finally, given that MDFU incurs a fixed deviation proportional to the message-loss rate, we adjust the accuracy of MDFU heuristically in a new protocol called MDFU with Linear Prediction (MDFU-LP). The evaluation shows that both MDFU and MDFU-LP behave very well in practice, even under high rates of message loss and even changing the input values dynamically.

cs.DC↗