We provide a comprehensive and updated assessment of Docker versus Docker in Docker (DinD), evaluating its impact on CPU, memory, disk, and network. Using different workloads, we evaluate DinD's performance across distinct hardware platforms and GNU/Linux distributions on cloud Infrastructure as a Service (laaS) platforms like Google Compute Engine (GCE) and traditional server-based environments. We developed an automated tools suite to achieve our goal. We execute four well-known benchmarks on Docker and its nested-container variant. Our findings indicate that nested-containers require up to 7 seconds for startup, while the Docker standard containers require less than 0.5 seconds for Debian and Alpine operating systems. Our results suggest that Docker containers based on Debian consistently outperform their Alpine counter-parts, showing lower CPU latency. A key distinction among these Docker images lies in the varying number of installed libraries (e.g., stretching from 13 to 119) across different Linux distributions for the same system (e.g., MySQL). Furthermore, the number of events and CPU latency indicates that the influence of DinD over Docker proves that it is insignificant for both operating systems. In terms of memory, running containers of Debian-based images consume 20% more size of memory than those based on Alpine. No significant differences are between nested-containers and Dockers for disk and network IO. It is worth emphasizing that some of the disparities, such as a bigger memory footprint, appear to be a direct result of the software stack in use, including different kernel versions. libraries. and other essential packages.

Monitoring principal entities in distributed systems is paramount to understanding the behavior of such systems and replicating their dynamics, e.g., for offline analysis or simulated evaluation. Examples of principal entities include users of distributed applications or online hosts on the internet. In many systems, monitoring can be done via periodic sampling of the entities online at a given instant. However, the monitoring process may be flawed. Some entities may fail to appear in one or more samples even though online when those samples were captured, thus compromising the quality of the monitoring traces. Previous investigations have applied statistical methods to identify such failures and used thresholds to correct them. In this paper, we turn our attention to machine learning methods as a means to regenerate monitoring traces collected via sampling. We propose a deep learning procedural method and two neural network topologies for correcting traces. We provide evidence that precision, accuracy, and recall can be substantially improved compared to existing statistical methods, thus opening an entire research avenue for improving the quality of monitoring traces of live distributed systems.

Content-Centric Networking (CCN) is a promising architectural approach that focuses on the efficient distribution of uniquely named data objects. A piece of content is represented by a single object in the network and is divided into multiple chunks which can be uniquely named and cached by network nodes. However, in its current form, the potential of CCN is not fully exploited due to the lack of common means to express and take advantage from possible relations that may exist among different objects. Our work explores the simple yet effective idea of supporting and exploiting such relations in CCN. In this paper, we present CCNrel as a backward-compatible mechanism for CCN that enables publishers to distribute contents as related objects. Differently from existing relation mechanisms, which focus on one type of content and are application-specific, CCNrel is generic and enables the use of relations in both current and novel application domains. First, we discuss CCNrel fundamental concepts and main design aspects. Next, we use CCNrel as foundation for a case study of data redundancy elimination in multimedia content distribution. Through extensive simulation work we evaluate the potential benefits of leveraging relations measured by the clients experience and overall network efficiency. Results of the presented use case show that, on average and when compared to default CCN operations, content download times are improved in 34%, publishers load in 56%, and the network bandwidth usage in 43%.

Recent analysis of the latest peer-to-peer trends worldwide indicates that BitTorrent is the most popular file sharing protocol, taking more than half of the P2P traffic in some geographical locations. Despite several studies about the dynamics of the “BitTorrent universe”, there exists no methodology to systematically observe it. This is mainly due to the challenges that need to be faced in order to observe the BitTorrent universe, the specificity of existing studies, and the ad hoc nature of the monitoring methods employed so far. In this paper, we propose a novel monitoring architecture (called TorrentU) that allows the systematic observation of large numbers of BitTorrent networks. Complementary monitoring strategies are flexibly combined to allow varying degrees of network/geographic coverage, information accuracy and richness of detail. To show the concept and technical feasibility of TorrentU, we implemented a prototype with the key parts of the architecture, and evaluated it through a case study with a rich set of monitoring campaigns running on PlanetLab nodes.

BitTorrent is probably the most popular file sharing protocol nowadays. Since it is a complex protocol which was created mostly as an engineering effort, there have been attempts to evaluate and understand the behavior of BitTorrent, exposing drawbacks and identifying opportunities for improvement. Simulation(al) and experimental evaluation are two important methodologies for the investigation of BitTorrent implementations and protocol variations, but using them with BitTorrent and P2P in general represents a challenge. In this paper, we introduce TorrentLab, a testbed in which BitTorrent simulations as well as live experiments can be performed under both controlled and uncontrolled settings. For a given swarm, we compare results obtained by means of simulation and live experiments, and show that both are in line with the expected behavior.

BitTorrent is a P2P file-sharing protocol that can be used to efficiently distribute files such as software updates and digital content to very large numbers of users. In a previous paper, we have shown that vulnerabilities can be exploited to launch Denial-of-Service attacks against BitTorrent swarms, which can substantially increase download times and network traffic. In this paper, we review the three most damaging attacks, and propose two algorithms as countermeasures to effectively tackle them. We implemented the attacks and countermeasures in a packet-level BitTorrent simulator. The results indicate that our proposed approach is effective when there is an ongoing attack while at the same time efficient when the countermeasure is active but there is no attack. To the best of our knowledge, this is the first proposal in the literature to make BitTorrent more robust against Denial-of-Service (DoS) attacks.

Tit-for-tat is widely believed to be the most effective strategy to enforce collaboration among selfish users. However, it has been shown that its usefulness for decentralized and dynamic environments such as peer-to-peer networks is marginal, as peers can rapidly end up in a deadlock situation. Many proposed solutions to this problem are either less resilient to freeloading behavior or induce a computational overhead that cannot be sustained by regular peers. In contrast, we retain tit-for-tat, but enhance the system with a novel form of source coding and an effective scheme to prevent peers from freeloading from seeding peers. We show that our system performs well without the risk of peer starvation and without sacrificing fairness. The proposed solution has a reasonably low overhead, and may hence be suitable for fully distributed content distribution applications in real networks.