A5061628886- Distributed systems and fault tolerance
- Peer-to-Peer Network Technologies
- Caching and Content Delivery
- Distributed and Parallel Computing Systems
- Advanced Software Engineering Methodologies
- Software System Performance and Reliability
- Privacy-Preserving Technologies in Data
- Cloud Computing and Resource Management
- Service-Oriented Architecture and Web Services
- Blockchain Technology Applications and Security
- Cryptography and Data Security
- Opportunistic and Delay-Tolerant Networks
- Advanced Data Storage Technologies
- Software Engineering Research
- Mobile Agent-Based Network Management
- Age of Information Optimization
- Mobile Ad Hoc Networks
- Recommender Systems and Techniques
- IoT and Edge/Fog Computing
- Cooperative Communication and Network Coding
- Complex Network Analysis Techniques
- Parallel Computing and Optimization Techniques
- Advanced Database Systems and Queries
- Optimization and Search Problems
- Real-Time Systems Scheduling
Université de Rennes
2014-2024
Institut de Recherche en Informatique et Systèmes Aléatoires
2014-2024
Centre National de la Recherche Scientifique
2002-2024
Institut national de recherche en informatique et en automatique
2011-2024
Inria Rennes - Bretagne Atlantique Research Centre
2012-2024
Université Rennes 2
2019
Lancaster University
2005-2013
Laboratoire d'Analyse et d'Architecture des Systèmes
2002-2005
Component-based software structuring principles are now commonplace at the application level; but componentization is far less established when it comes to building low-level systems software. Although there have been pioneering efforts in applying systems-building, these tended target specific domains (e.g., embedded systems, operating communications programmable networking environments, or middleware platforms). They also tend be targeted deployment environments standard personal computer...
In order to provide an increasing number of functionalities and benefit from sophisticated application-tailored services the network, distributed applications are led integrate ever-widening range networking technologies. As these become more complex, this requirement for 'network heterogeneity' is becoming a crucial issue in their development. Although progress has been made community addressing such needs through development network overlays, we claim paper that middleware slow advances...
During the last decade, Machine Learning (ML) has increasingly become a hot topic in field of Computer Networks and is expected to be gradually adopted for plethora control, monitoring management tasks real-world deployments. This poses need count on new generations students, researchers practitioners with solid background ML applied networks. 2020, International Telecommunication Union (ITU) organized "ITU AI/ML 5G challenge", an open global competition that introduced broad audience some...
Federated learning (FL) is very appealing for its privacy benefits: essentially, a global model trained with updates computed on mobile devices while keeping the data of users local. Standard FL infrastructures are however designed to have no energy or performance impact devices, and therefore not suitable applications that require frequent ( online ) updates, such as news recommenders. This article presents FLeet , first Online system, acting middleware between Android operating system...
Coupling metrics received increased recognition by object-oriented (OO) software developers when they were found to be indicators of important quality attributes, such as fault-proneness. However, there is no consensus on which coupling are effective for emerging development paradigms, Aspect-Oriented Programming (AOP). AOP aims improve providing significantly different decomposition mechanisms, pointcut, advice and intertype declarations. Therefore, it not obvious if can derived from direct...
User Generated Content (UGC), such as YouTube videos, accounts for a substantial fraction of the Internet traffic. To optimize their performance, UGC services usually rely on both proactive and reactive approaches that exploit spatial temporal locality in access patterns. Alternative types are also relevant hardly ever considered together. In this paper, we show large (more than 650,000 videos) dataset content (induced by related videos feature) geographic locality, fact correlated. More...
The world's computing infrastructure is increasingly differentiating into self-contained distributed systems with various purposes and capabilities (e.g. IoT installations, clouds, VANETs, WSNs, CDNs, . .). Furthermore, such are being composed to generate of that offer value-added functionality. Today, however, system composition typically ad-hoc fragile. It requires developers possess an intimate knowledge internals low-level interactions between their components. In this paper, we outline...
To preserve the Bitcoin ledger's integrity, a node that joins system must download full copy of entire blockchain if it wants to verify newly created blocks.
Coupling in software applications is often used as an indicator of external quality attributes such fault-proneness. In fact, the correlation coupling metrics and faults object oriented programs has been widely studied. However, there very limited knowledge about which properties aspect-oriented programming (AOP) are effective indicators modules. Existing do not take into account specificities AOP mechanisms. As a result, these unlikely to provide optimal predictions pivotal This impacts...
Erasure coding is an established data protection mechanism. It provides high resiliency with low storage overhead, which makes it very attractive to systems developers. Unfortunately, when used in a distributed setting, erasure hampers system's performance, because requires clients contact several, possibly remote sites retrieve their data. This has hindered the adoption of practice, limiting its use cold, archival Recent research showed that feasible for hot as well, thus opening new...
The Middleware conference's mission is to facilitate the development and implementation of distributed systems, whether mobile, cloudbased, or geographically distributed. conference focuses on design principles, programming models, frameworks, runtime analysis support. This year's continues long tradition bringing together academic industrial participants for vibrant discussions presentations cutting-edge research.
K-Nearest-Neighbor (KNN) graphs have emerged as a fundamental building block of many on-line services providing recommendation, similarity search and classification. Constructing KNN graph rapidly accurately is, however, computationally intensive task. As data volumes keep growing, speed the ability to scale out are becoming critical factors when deploying algorithm. In this work, we present KIFF, generic, fast scalable construction KIFF directly exploits bipartite nature most datasets which...
Most recent CRDT techniques rely on a causal broadcast primitive to provide guarantees the delivery of operation deltas. Such is unfortunately hard implement efficiently in large open networks, whose membership often difficult track. As an alternative, we argue this paper that pure state-based CRDTs can be implemented by encoding states as specialized Merkle trees, and approach well suited networks where many nodes may join leave. At core our contribution lies new kind tree, called Search...
Geolocated social networks, combining traditional networking features with geolocation information, have grown tremendously over the last few years. Yet, very works looked at implementing geolocated networks in a fully distributed manner, promising avenue to handle growing scalability challenges of these systems. In this paper, we focus on georecommendation, and show that existing decentralized recommendation mechanisms perform fact poorly geodata. We propose set novel gossip-based address...
This paper considers distributed vertex-coloring in broadcast/receive networks suffering from conflicts and collisions. (A collision occurs when, during the same round, messages are sent to process by too many neighbors; a conflict when one of its neighbors broadcast round.) More specifically, focuses on multi-channel networks, which may either message or receive at most γ them. The first provides new upper bound corresponding graph coloring problem (known as frugal coloring) general graphs,...
Peer-to-peer (P2P) systems have been widely researched over the past decade, leading to highly scalable implementations for a wide range of distributed services and applications. A P2P system assigns symmetric roles machines, which can act both as client server. This alleviates need any central component maintain global knowledge system. Instead, each peer takes individual decisions based on local rest system, providing scalability by design. While successfully applied applications...
Developers must often diagnose anomalies in programs they only have a partial knowledge of. As result, simultaneously reverse engineer parts of the system are unfamiliar with while interpreting dynamic observation data (performance profiling traces, error-propagation channels, memory leaks), task particularly difficult. To support developers this kind comprehension task, filtering and aggregation long been suggested as key enabling strategies. Unfortunately, traditional approaches typically...
Aspect-oriented programming (AOP) aims at facilitating program comprehension and maintenance in the presence of crosscutting concerns. Aspect code is often introduced extended as software projects evolve. Unfortunately, we still lack a good understanding how faults are evolving aspect-oriented programs. More importantly, there little knowledge whether existing metrics related to typical fault introduction processes code. This paper presents an exploratory study focused on analysis during...
The past few years have seen an increasingly tight link between grid computing and Web services, with the latest standards defining a architecture as set of services built using protocols. However, reputation these technologies (SOAP, XML, WSDL, HTTP) is that they are heavyweight slow, something potentially concern given current anticipated application mix for high performance architectures. This paper reports results evaluation carried out on Globus 3.9.4, reference implementation new GGF...