Distributed Applications and Interoperable Systems

Byzantine fault tolerant (BFT) applications are usually implemented with dedicated clients that interact with a set of replicas with some BFT protocol. In this paper, we explore the possibility of using web-based clients for interaction with a BFT service. Our contributions address the trustworthy deployment of client code and configuration in a browser-based execution environment (client bootstrapping), the design and implementation of a BFT client within the constraints of a browser-based JavaScript execution environment, and publish-subscribe extensions to the standard request/reply interaction model of BFT state machine model to simplify the implementation of efficient interactive web applications.

This paper outlines work on the detection of anomalous behaviour in Online Social Networks (OSNs). We present various automated techniques for identifying a ‘prodigious’ segment within a tweet, and consider tweets which are unusual because of writing style, posting sequence, or engagement level. We evaluate the mechanism by running extensive experiments over large artificially constructed tweets corpus, crawled to include randomly interpolated and abnormal Tweets. In order to successfully identify anomalies in a tweet, we aggregate more than 21 features to characterize users’ behavioural pattern. Using these features with each of our methods, we examine the effect of the total number of tweets on our ability to detect an anomaly, allowing segments of size 50 tweets 100 tweets and 200 tweets. We show indispensable improvements over a baseline in all circumstances for each method, and identify the method variant which performs persistently better than others.

Mobile applications tend to ask for users’ location in order to improve the service they provide. However, aside from increasing their service utility, they may also store these data, analyze them or share them with external parties. These privacy threats for users are a hot topic of research, leading to the development of so called Location Privacy Protection Mechanisms. LPPMs often are configurable algorithms that enable the tuning of the privacy protection they provide and thus the leveraging of the service utility. However, they usually do not provide ways to measure the achieved privacy in practice for all users of mobile devices, and even less clues on how a given configuration will impact privacy of the data given the specificities of everyone’s mobility. Moreover, as most Location Based Services require the user position in real time, these measures and predictions should be achieved in real time. In this paper we present a metric to evaluate privacy of obfuscated data based on users’ points of interest as well as a predictive model of the impact of a LPPM on these measure; both working in a real time fashion. The evaluation of the paper’s contributions is done using the state of the art LPPM Geo-I on synthetic mobility data generated to be representative of real-life users’ movements. Results highlight the relevance of the metric to capture privacy, the fitting of the model to experimental data, and the feasibility of the on-line mechanisms due to their low computing complexity.

Handling workloads generated by a large number of users, data-stream–processing systems also require large amounts of energy. To reduce their energy footprint, such systems typically rely on the operating systems of their servers to adjust processor speeds depending on the current workload by performing dynamic voltage and frequency scaling (DVFS). In this paper, we show that, although effective, this approach still leaves room for significant energy savings due to DVFS making conservative assumptions regarding its impact on application performance. To leverage the unused potential we present Strome, an energy-aware technique to minimize energy demand in data-stream–processing systems by dynamically adapting upper limits for the power demand of hardware components. In contrast to DVFS, Strome exploits information on application performance and is therefore able to achieve energy savings while minimizing its effects on throughput and latency. Our evaluation shows that Strome is particularly effective in the face of varying workloads, reducing power demand by up to 25 % compared with the state-of-the-art data-stream–processing system Heron relying on DVFS.

Scalability is one of the most relevant features of today’s data management systems. In order to achieve high scalability and availability, recent distributed key-value stores refrain from costly replica coordination when processing requests. However, these systems typically do not perform well under churn. In this paper, we propose DataFlagons, a large-scale key-value store that integrates epidemic dissemination with a probabilistic total order broadcast algorithm. By ensuring that all replicas process requests in the same order, DataFlagons provides probabilistic strong data consistency while achieving high scalability and robustness under churn.

We render a model advocating an extension of choreographies to describe reverse computation via monitoring. More precisely, our extension imbues the communication behaviour of multi-party protocols with minimal decorations specifying the conditions triggering monitor adaptations. We show how, from these extended global descriptions, one can (i) synthesise actors implementing the normal local behaviour of the system prescribed by the global graph, but also (ii) synthesise monitors that are able to coordinate a distributed rollback when certain conditions (denoting abnormal behaviour) are met.

The study of complex networks has acquired great importance during the last years because of the diffusion of several phenomena which can be described by these networks. Community detection is one of the most investigated problem in this area, however only a few solutions for detecting communities in a distributed and dynamic environment have been presented. In this paper we propose SONIC-MAN, a distributed protocol to detect and manage communities in a peer-to-peer dynamic environment. Our approach is particularly targeted to distributed online social networks and its main goal is to discover communities in the ego-network of the users. SONIC-MAN is based on a Temporal Trade-off approach and exploits a set of super-peers for the management of the communities. The paper presents a set of evaluations proving that SONIC-MAN is able to detect dynamic communities in a distributed setting and to return results close a centralized approach based on the same basic algorithm for community discovering.

Messaging applications are among the most popular internet applications and people use them worldwide on a daily basis. Their supporting infrastructure, though consisting of a multitude of servers, is typically under central control. This enables censorship and seamless user profiling. A fully decentralized infrastructure, with decentralized control and redundant data storage, can mitigate these dangers. In this paper we evaluate the basic ability of decentralized networks created by the network overlay and data storage protocol Kademlia to serve as a short-term data cache for messaging applications. Our results show, that reliable retrieval of up to 20 replicas is possible.

Finding the most efficient policy for the association of objects with threads is one of the main challenges in the deployment of concurrently executing objects, including actors. For the case of actor-based programs, libraries, frameworks, and languages provide fine tuning facilities for associating actors with threads. In practice, programmers use the default policy for the initial deployment of actors and the default policy is replaced with some other policies considering runtime behaviors of actors. Although this ad-hoc approach is widely used by programmers, it is tedious and time-consuming for large-scale applications. To reduce the time-consumption of the ad-hoc approach, a set of heuristics is proposed with the aim of balancing computations of actors across threads. This technique results in performance improvement; however, it relies on the static analysis of source codes and actors’ behaviors, ends in the inappropriate configuration of systems in distributed environments. In this paper, we illustrate conditions that the proposed heuristics do not work well and propose a new approach based on the runtime profile of actors for better association of actors with threads. We also show how this approach can be extended to a fully self-adaptive approach and illustrated its applicability using a set of case studies.

We describe a case study application for untethered video dissemination using a hybrid edge-cloud architecture featuring Android devices, possibly organised in WiFi-Direct groups, and Raspberry Pi-based cloudlets, structured in a mesh and also working as access points. The application was tested in the real-world scenario of a Portuguese volleyball league game. During the game, users of the application recorded videos and injected them in the edge-cloud. The cloudlet servers continuously synchronised their cached video contents over the mesh network, allowing users on different locations to share their videos, without resorting to any other network infrastructure. An analysis of the logs gathered during the experiment shows that such portable setups can easily disseminate videos to tens of users through the edge-cloud with low latencies. We observe that the edge-cloud may be naturally resilient to faulty cloudlets or devices, taking advantage of video caching within devices and WiFi-Direct groups, and of device churn to opportunistically disseminate videos.

Today, many users of multimedia applications are surrounded by a changing set of multimedia-capable devices. However, users can move their running multimedia applications only to a pre-defined set of devices. Application mobility is the paradigm where users can move their running applications (or parts of) to heterogeneous devices in a seamless manner. In order to continue multimedia processing under the implied context changes in application mobility, applications need to adapt the presentation of multimedia content and their internal configuration. We propose the system DAMPAT that implements an adaptation control loop to adapt multimedia pipelines. Exponential combinatorial growth of possible pipeline configurations is controlled by architectural constraints specified as high-level goals by application developers. Our evaluation shows that the pipeline only needs to be interrupted a few tens of milliseconds to perform the reconfiguration. Thus, production or consumption of multimedia content can continue across heterogeneous devices and user context changes in a seamless manner.

The microservice architecture is a subtype of distributed systems that has been attracting a lot of interest both in the industry and academia. Emerging research recognizes the need for a better understanding of microservice security, and, in particular, mechanisms that enable defense-in-depth and adaptive security. With the continuously growing number of possible attacks and defenses, the choice in the optimal defense strategy becomes non-trivial as well as time critical. We propose a cost-sensitive adaptable intrusion response system for microservices, which uses a game theoretic approach to automatically respond to network attacks in real time. Furthermore, we discuss both the applicable attacks and defense responses specific to microservices.