arXiv subjects

Andrea Passarella

Publications and source records attributed to Andrea Passarella.

At least 91 records · Page 5Linked to original sources

arXiv2022

On Efficiently Partitioning a Topic in Apache Kafka

Apache Kafka addresses the general problem of delivering extreme high volume event data to diverse consumers via a publish-subscribe messaging system. It uses partitions to scale a topic across many brokers for producers to write data in parallel, and also to facilitate parallel reading of consumers. Even though Apache Kafka provides some out of the box optimizations, it does not strictly define how each topic shall be efficiently distributed into partitions. The well-formulated fine-tuning that is needed in order to improve an Apache Kafka cluster performance is still an open research problem. In this paper, we first model the Apache Kafka topic partitioning process for a given topic. Then, given the set of brokers, constraints and application requirements on throughput, OS load, replication latency and unavailability, we formulate the optimization problem of finding how many partitions are needed and show that it is computationally intractable, being an integer program. Furthermore, we propose two simple, yet efficient heuristics to solve the problem: the first tries to minimize and the second to maximize the number of brokers used in the cluster. Finally, we evaluate its performance via large-scale simulations, considering as benchmarks some Apache Kafka cluster configuration recommendations provided by Microsoft and Confluent. We demonstrate that, unlike the recommendations, the proposed heuristics respect the hard constraints on replication latency and perform better w.r.t. unavailability time and OS load, using the system resources in a more prudent way.

cs.NI↗

arXiv2022

Modeling Human Behavior Part II -- Cognitive approaches and Uncertainty

As we discussed in Part I of this topic, there is a clear desire to model and comprehend human behavior. Given the popular presupposition of human reasoning as the standard for learning and decision-making, there have been significant efforts and a growing trend in research to replicate these innate human abilities in artificial systems. In Part I, we discussed learning methods which generate a model of behavior from exploration of the system and feedback based on the exhibited behavior as well as topics relating to the use of or accounting for beliefs with respect to applicable skills or mental states of others. In this work, we will continue the discussion from the perspective of methods which focus on the assumed cognitive abilities, limitations, and biases demonstrated in human reasoning. We will arrange these topics as follows (i) methods such as cognitive architectures, cognitive heuristics, and related which demonstrate assumptions of limitations on cognitive resources and how that impacts decisions and (ii) methods which generate and utilize representations of bias or uncertainty to model human decision-making or the future outcomes of decisions.

cs.AI↗

arXiv2022

Modeling Human Behavior Part I -- Learning and Belief Approaches

There is a clear desire to model and comprehend human behavior. Trends in research covering this topic show a clear assumption that many view human reasoning as the presupposed standard in artificial reasoning. As such, topics such as game theory, theory of mind, machine learning, etc. all integrate concepts which are assumed components of human reasoning. These serve as techniques to attempt to both replicate and understand the behaviors of humans. In addition, next generation autonomous and adaptive systems will largely include AI agents and humans working together as teams. To make this possible, autonomous agents will require the ability to embed practical models of human behavior, which allow them not only to replicate human models as a technique to "learn", but to to understand the actions of users and anticipate their behavior, so as to truly operate in symbiosis with them. The main objective of this paper it to provide a succinct yet systematic review of the most important approaches in two areas dealing with quantitative models of human behaviors. Specifically, we focus on (i) techniques which learn a model or policy of behavior through exploration and feedback, such as Reinforcement Learning, and (ii) directly model mechanisms of human reasoning, such as beliefs and bias, without going necessarily learning via trial-and-error.

cs.AI↗

arXiv2022

Stateless or stateful FaaS? I'll take both!

Serverless computing has emerged as a very popular cloud technology, together with its companion Function-as-a-Service (FaaS) programming model enabling invocations of stateless functions from clients. An evolution of serverless is now taking place, shifting it towards the edge of the network and broadening its scope to stateful functions, as well. In this paper we argue that stateless vs. stateful is not a dichotomy of the application per se, but rather a time-varying property of most (if not all) applications, as confirmed by the analysis of real traces collected in a production environment. Based on this observation, we propose a mathematical formulation of a resource allocation problem that jointly encompasses both operation modes, dubbed lambda vs. mu, which can be solved efficiently at run-time by an edge orchestrator. We evaluate the proposed solution via simulation experiments in realistic network and workload conditions, which leads the way to the practical realization of a system where applications can freely adapt their current operation mode and optimize their performance at a minimum cost of operation from the network's perspective.

cs.NI↗

arXiv2022

Wireless Crowd Charging with Battery Aging Mitigation

Battery aging is one of the major concerns for the pervasive devices such as smartphones, wearables and laptops. Current battery aging mitigation approaches only partially leverage the available options to prolong battery lifetime. In this regard, we claim that wireless crowd charging via network-wide smart charging protocols can provide a useful setting for applying battery aging mitigation. In this paper, for the first time in the state-of-the-art, we couple the two concepts and we design a fine-grained battery aging model in the context of wireless crowd charging, and two network-wide protocols to mitigate battery aging. Our approach directly challenges the related contemporary research paradigms by (i) taking into account important characteristic phenomena in the algorithmic modeling process related to fine-grained battery aging properties, (ii) deploying ubiquitous computing and network-wide protocols for battery aging mitigation, and (iii) fulfilling the user QoE expectations with respect to the enjoyment of a longer battery lifetime. Simulation-based results indicate that the proposed protocols are able to mitigate battery aging quickly in terms of nearly 46.74-60.87% less reduction of battery capacity among the crowd, and partially outperform state-of-the-art protocols in terms of energy balance quality.

cs.NI↗

arXiv2022

Resource Allocation in Quantum Networks for Distributed Quantum Computing

The evolution of quantum computing technologies has been advancing at a steady pace in the recent years, and the current trend suggests that it will become available at scale for commercial purposes in the near future. The acceleration can be boosted by pooling compute infrastructures to either parallelize algorithm execution or solve bigger instances that are not feasible on a single quantum computer, which requires an underlying Quantum Internet: the interconnection of quantum computers by quantum links and repeaters to exchange entangled quantum bits. However, Quantum Internet research so far has been focused on provisioning point-to-point flows only, which is suitable for (e.g.) quantum sensing and metrology, but not for distributed quantum computing. In this paper, after a primer on quantum computing and networking, we investigate the requirements and objectives of smart computing on distributed nodes from the perspective of quantum network provisioning. We then design a resource allocation strategy that is evaluated through a comprehensive simulation campaign, whose results highlight the key features and performance issues, and lead the way to further investigation in this direction.

quant-ph↗

arXiv2022

Reliable data delivery in ICN-IoT environments

In an IoT environment, which is characterized by a multitude of interconnected smart devices with sensing and computational capabilities, many applications are (i) content-based, that is, they are only interested in finding a given type of content rather than the location of data, and (ii) contextualized, that is, the content is generated and consumed in the proximity of the user. In this context, the Information-Centric Networking (ICN) paradigm is an appealing model for efficiently retrieving application data, and the service decentralization towards the network edge helps to reduce the core network load being the data produced by IoT devices mainly confined in the area where they are generated. MobCCN is an ICN-based data delivery protocol that we designed for operating efficiently in such context [1][2], where static and mobile IoT devices are enriched with ICN functions. Specifically, MobCCN leverages an efficient routing and forwarding protocol, exploiting opportunistic contacts among IoT mobile devices, to fill the Forwarding Interest Base (FIB) tables so as to correctly forward Interest packets towards the intended data producers. In this paper, we aim to enhance the reliability of MobCCN by exploring different retransmissions mechanisms, such as retransmissions based on number of duplicate Interests that are received for the same requested content, periodic retransmissions, single path versus disjoint multi-path forwarding, hysteresis mechanism and combinations of them. Extensive simulation results show that, among the analysed MobCCN variants, the one that implements both periodic retransmissions and a hysteresis-based retransmission process ensures the highest delivery rates (up to 95\%) and the shortest network path, with a very limited traffic overhead due to retransmissions.

cs.NI↗

arXiv2022

Balanced Wireless Crowd Charging with Mobility Prediction and Social Awareness

The advancements in peer-to-peer wireless power transfer (P2P-WPT) have empowered the portable and mobile devices to wirelessly replenish their battery by directly interacting with other nearby devices. The existing works unrealistically assume the users to exchange energy with any of the users and at every such opportunity. However, due to the users' mobility, the inter-node meetings in such opportunistic mobile networks vary, and P2P energy exchange in such scenarios remains uncertain. Additionally, the social interests and interactions of the users influence their mobility as well as the energy exchange between them. The existing P2P-WPT methods did not consider the joint problem for energy exchange due to user's inevitable mobility, and the influence of sociality on the latter. As a result of computing with imprecise information, the energy balance achieved by these works at a slower rate as well as impaired by energy loss for the crowd. Motivated by this problem scenario, in this work, we present a wireless crowd charging method, namely MoSaBa, which leverages mobility prediction and social information for improved energy balancing. MoSaBa incorporates two dimensions of social information, namely social context and social relationships, as additional features for predicting contact opportunities. In this method, we explore the different pairs of peers such that the energy balancing is achieved at a faster rate as well as the energy balance quality improves in terms of maintaining low energy loss for the crowd. We justify the peer selection method in MoSaBa by detailed performance evaluation. Compared to the existing state-of-the-art, the proposed method achieves better performance trade-offs between energy-efficiency, energy balance quality and convergence time.

cs.NI↗

arXiv2021

Stateful Function-as-a-Service at the Edge

In FaaS, users invoke remote functions, which encapsulate service(s). These functions typically need to remotely access a persistent state via external services: this makes the paradigm less attractive in edge systems, especially for IoT applications, due to the increased delay and outbound traffic. We propose to generalize the FaaS paradigm by allowing functions to alternate between remote-state and local-state phases, depending on internal and external conditions, and dedicating a container with persistent memory to functions when in a local-state phase. We present initial results showing that this simple yet powerful pattern allows to better utilize the available resources, which are scarce on edge nodes, while significantly reducing tail latencies, which is key to enable many new applications based on real-time ML, e.g., in smart vehicles and smart factory scenarios

cs.NI↗

arXiv2021

Predictive Auto-scaling with OpenStack Monasca

Cloud auto-scaling mechanisms are typically based on reactive automation rules that scale a cluster whenever some metric, e.g., the average CPU usage among instances, exceeds a predefined threshold. Tuning these rules becomes particularly cumbersome when scaling-up a cluster involves non-negligible times to bootstrap new instances, as it happens frequently in production cloud services. To deal with this problem, we propose an architecture for auto-scaling cloud services based on the status in which the system is expected to evolve in the near future. Our approach leverages on time-series forecasting techniques, like those based on machine learning and artificial neural networks, to predict the future dynamics of key metrics, e.g., resource consumption metrics, and apply a threshold-based scaling policy on them. The result is a predictive automation policy that is able, for instance, to automatically anticipate peaks in the load of a cloud application and trigger ahead of time appropriate scaling actions to accommodate the expected increase in traffic. We prototyped our approach as an open-source OpenStack component, which relies on, and extends, the monitoring capabilities offered by Monasca, resulting in the addition of predictive metrics that can be leveraged by orchestration components like Heat or Senlin. We show experimental results using a recurrent neural network and a multi-layer perceptron as predictor, which are compared with a simple linear regression and a traditional non-predictive auto-scaling policy. However, the proposed framework allows for the easy customization of the prediction policy as needed.

cs.DC↗

arXiv2021

Dynamic Hard Pruning of Neural Networks at the Edge of the Internet

Neural Networks (NN), although successfully applied to several Artificial Intelligence tasks, are often unnecessarily over-parametrised. In edge/fog computing, this might make their training prohibitive on resource-constrained devices, contrasting with the current trend of decentralising intelligence from remote data centres to local constrained devices. Therefore, we investigate the problem of training effective NN models on constrained devices having a fixed, potentially small, memory budget. We target techniques that are both resource-efficient and performance effective while enabling significant network compression. Our Dynamic Hard Pruning (DynHP) technique incrementally prunes the network during training, identifying neurons that marginally contribute to the model accuracy. DynHP enables a tunable size reduction of the final neural network and reduces the NN memory occupancy during training. Freed memory is reused by a \emph{dynamic batch sizing} approach to counterbalance the accuracy degradation caused by the hard pruning strategy, improving its convergence and effectiveness. We assess the performance of DynHP through reproducible experiments on three public datasets, comparing them against reference competitors. Results show that DynHP compresses a NN up to $10$ times without significant performance drops (up to $3.5\%$ additional error w.r.t. the competitors), reducing up to $80\%$ the training memory occupancy.

cs.LG↗

arXiv2021

A Decentralized Framework for Serverless Edge Computing in the Internet of Things

Serverless computing is becoming widely adopted among cloud providers, thus making increasingly popular the Function-as-a-Service (FaaS) programming model, where the developers realize services by packaging sequences of stateless function calls. The current technologies are very well suited to data centers, but cannot provide equally good performance in decentralized environments, such as edge computing systems, which are expected to be typical for Internet of Things (IoT) applications. In this paper, we fill this gap by proposing a framework for efficient dispatching of stateless tasks to in-network executors so as to minimize the response times while exhibiting short- and long-term fairness, also leveraging information from a virtualized network infrastructure when available. Our solution is shown to be simple enough to be installed on devices with limited computational capabilities, such as IoT gateways, especially when using a hierarchical forwarding extension. We evaluate the proposed platform by means of extensive emulation experiments with a prototype implementation in realistic conditions. The results show that it is able to smoothly adapt to the mobility of clients and to the variations of their service request patterns, while coping promptly with network congestion.

cs.NI↗

arXiv2021

Structural invariants in individuals language use: the "ego network" of words

The cognitive constraints that humans exhibit in their social interactions have been extensively studied by anthropologists, who have highlighted their regularities across different types of social networks. We postulate that similar regularities can be found in other cognitive processes, such as those involving language production. In order to provide preliminary evidence for this claim, we analyse a dataset containing tweets of a heterogeneous group of Twitter users (regular users and professional writers). Leveraging a methodology similar to the one used to uncover the well-established social cognitive constraints, we find that a concentric layered structure (which we call ego network of words, in analogy to the ego network of social relationships) very well captures how individuals organise the words they use. The size of the layers in this structure regularly grows (approximately 2-3 times with respect to the previous one) when moving outwards, and the two penultimate external layers consistently account for approximately 60% and 30% of the used words (the outermost layer contains 100% of the words), irrespective of the number of the total number of layers of the user.

cs.SI↗

arXiv2021

Dynamics of scientific collaboration networks due to academic migrations

Academic migration is the change of host institution by a researcher, typically aimed at achieving a stronger research profile. Scientific features such as the number of collaborations, the productivity and its research impact tend to be directly affected by such movements. In this paper, we analyse the dynamics of the collaboration network of researchers as they move from an institution to the next one. We specifically highlight cases where they increase and when they shrink, and quantify the dependency between the collaboration networks before and after such a movement. Finally, we drill down the analysis by dividing movements depending on the career stage of the researchers. The analysis shows a remarkable dynamism of collaboration networks across migrations. Interestingly, not always movements result in larger collaboration networks, while the overall similarity between networks across movements is quite limited on average. Qualitatively, the same effects can be found at all career stages, while, clearly, the magnitude of them might vary. These results are based on a dataset extracted from Scopus, containing detailed scientific information for the publications of 84,141 researchers.

cs.DL↗

arXiv2021

Cellular traffic offloading via Opportunistic Networking with Reinforcement Learning

The widespread diffusion of mobile phones is triggering an exponential growth of mobile data traffic that is likely to cause, in the near future, considerable traffic overload issues even in last-generation cellular networks. Offloading part of the traffic to other networks is considered a very promising approach and, in particular, in this paper, we consider offloading through opportunistic networks of users' devices. However, the performance of this solution strongly depends on the pattern of encounters between mobile nodes, which should therefore be taken into account when designing offloading control algorithms. In this paper, we propose an adaptive offloading solution based on the Reinforcement Learning framework and we evaluate and compare the performance of two well-known learning algorithms: Actor-Critic and Q-Learning. More precisely, in our solution the controller of the dissemination process, once trained, is able to select a proper number of content replicas to be injected into the opportunistic network to guarantee the timely delivery of contents to all interested users. We show that our system based on Reinforcement Learning is able to automatically learn a very efficient strategy to reduce the traffic on the cellular network, without relying on any additional context information about the opportunistic network. Our solution achieves a higher level of offloading with respect to other state-of-the-art approaches, in a range of different mobility settings. Moreover, we show that a more refined learning solution, based on the Actor-Critic algorithm, is significantly more efficient than a simpler solution based on Q-learning.

cs.NI↗

arXiv2021

A Social Cognitive Heuristic for Adaptive Data Dissemination in Mobile Opportunistic Networks

It is commonly agreed that data will be one of the cornerstones of Future Internet systems. In this context, mobile Opportunistic Networks (ONs) are one of the key paradigms to support, in a self-organising and decentralised manner, the growth of data generated by localized interactions between users mobile devices, and between them and nearby devices such as IoT nodes. In ONs, the spontaneous collaboration among mobile devices is exploited to disseminate data toward interested users. However, the limited resources and knowledge available at each node, and the vast amount of data available, make it difficult to devise efficient schemes to accomplish this task. Recent solutions propose to equip each device with data filtering methods derived from human data processing schemes, known as Cognitive Heuristics, i.e. very effective methods used by the brain to quickly drop useless information, while keeping the most relevant one. These solutions can become less effective when facing dynamic scenarios or situations where nodes cannot fully collaborate. One of the reasons is that the solutions proposed so far do not take take into account the social structure of the environment where the nodes move in. To be more effective, the selection of information performed by each node should take into consideration this dimension of the environment. In this paper we propose a social-based data dissemination scheme, based on the cognitive Social Circle Heuristic. This evaluation method exploits the structure of the social environment to make inferences about the relevance of discovered information. We show how the Social Circle Heuristic, coupled with a cognitive-based community detection scheme, can be exploited to design an effective data dissemination algorithm for ONs. We provide a detailed analysis of the performance of the proposed solution via simulation.

cs.NI↗

arXiv2021

Human-centric Data Dissemination in the IoP: Large-scale Modeling and Evaluation

Data management using Device-to-Device (D2D) communications and opportunistic networks (ONs) is one of the main focuses of human-centric pervasive Internet services. In the recently proposed "Internet of People" paradigm, accessing relevant data dynamically generated in the environment nearby is one of the key services. Moreover, personal mobile devices become proxies of their human users while exchanging data in the cyber world and, thus, largely use ONs and D2D communications for exchanging data directly. Recently, researchers have successfully demonstrated the viability of embedding human cognitive schemes in data dissemination algorithms for ONs. In this paper, we consider one such scheme based on the recognition heuristic, a human decision-making scheme used to efficiently assess the relevance of data. While initial evidence about its effectiveness is available, the evaluation of its behaviour in large-scale settings is still unsatisfactory. To overcome these limitations, we have developed a novel hybrid modelling methodology, which combines an analytical model of data dissemination within small-scale communities of mobile users, with detailed simulations of interactions between different communities. This methodology allows us to evaluate the algorithm in large-scale city- and country-wide scenarios. Results confirm the effectiveness of cognitive data dissemination schemes, even when content popularity is very heterogenous.

cs.NI↗

arXiv2021

A communication efficient distributed learning framework for smart environments

Due to the pervasive diffusion of personal mobile and IoT devices, many ``smart environments'' (e.g., smart cities and smart factories) will be, among others, generators of huge amounts of data. Currently, this is typically achieved through centralised cloud-based data analytics services. However, according to many studies, this approach may present significant issues from the standpoint of data ownership, and even wireless network capacity. One possibility to cope with these shortcomings is to move data analytics closer to where data is generated. In this paper, we tackle this issue by proposing and analyzing a distributed learning framework, whereby data analytics are performed at the edge of the network, i.e., on locations very close to where data is generated. Specifically, in our framework, partial data analytics are performed directly on the nodes that generate the data, or on nodes close by (e.g., some of the data generators can take this role on behalf of subsets of other nodes nearby). Then, nodes exchange partial models and refine them accordingly. Our framework is general enough to host different analytics services. In the specific case analysed in the paper, we focus on a learning task, considering two distributed learning algorithms. Using an activity recognition and a pattern recognition task, both on reference datasets, we compare the two learning algorithms between each other and with a central cloud solution (i.e., one that has access to the complete datasets). Our results show that using distributed machine learning techniques, it is possible to drastically reduce the network overhead, while obtaining performance comparable to the cloud solution in terms of learning accuracy. The analysis also shows when each distributed learning approach is preferable, based on the specific distribution of the data on the nodes.

cs.DC↗