2023 Conference article Open Access

See at: ISTI Repository | ieeexplore.ieee.org | CNR ExploRA

2023 Conference article Open Access

DATA7: a dataset for assessing resource and application management solutions at the edge
Carlini E., Coppola M., Dazzi P., Ferrucci L., Kavalionak H., Mordacchini M.
This paper presents a dataset on edge devices and mobility patterns to comprehensively understand user behaviour and devices workload in Edge computing environments. The dataset is built on top of a publicly available dataset of cellular tower locations to simulate Edge devices, and on user mobility trajectories generated by a state-of-the-art simulator based on real location maps in the area of the city of Pisa, Italy. The resulting dataset reports the amount of vehicles in the range of about 200 Edge devices for each step of the simulation. The dataset can be used for various applications in edge computing and mobility, most notably for assessing results on resource and application management solutions at the edge in a realistic environment.Source: FRAME '23 - 3rd Workshop on Flexible Resource and Application Management on the EdgeAugust 2023 - colocated with HPDC '23 - 32nd International Symposium on High-Performance Parallel and Distributed Computing, pp. 3–6, Orlando, Florida, USA, 20/06/2023
DOI: 10.1145/3589010.3595652
Project(s): ACCORDION
Metrics:

See at: ISTI Repository | dl.acm.org | CNR ExploRA

2023 Journal article Open Access

See at: doi.org | ISTI Repository | www.frontiersin.org | CNR ExploRA

2023 Conference article Open Access

TEACHING: A Computing Toolkit for Building Efficient Autonomous appliCations Leveraging Humanistic INtelliGence
Bacciu D., Tserpes K., Coppola M., Macher G., Gallicchio C., Veledar O., Anaxagorou A. M., Dazzi P.
TEACHING proposes a distributed, trustworthy AI integrating continuous human feedback, supporting CPSoS application design and deployment. TEACHING envisions an intelligent environment, empowering humans through cybernetic assistance. It advances autonomous safety-critical systems, improving safety, reliability and acceptability through human-centred design and formal validation crossing paradigms. TEACHING brings humans and AI together, enabling participatory development, optimisation and oversight.Source: FRAME '23: 3rd Workshop on Flexible Resource and Application Management on the Edge, pp. 37–39, Orlando, USA, 20/06/2023
DOI: 10.1145/3589010.3594886
Project(s): TEACHING
Metrics:

See at: dl.acm.org | ISTI Repository | doi.org | CNR ExploRA

2022 Conference article Open Access

A novel approach to distributed model aggregation using Apache Kafka
Bano S., Carlini E., Cassarà P., Coppola M., Dazzi P., Gotta A.
Multi-Access Edge Computing (MEC) is attracting a lot of interest because it complements cloud-based approaches. Indeed, MEC is opening up in the direction of reducing both interaction delays and data sharing, called Cyber-Physical Systems (CPSs). In the near fu-ture, edge technologies will be a fundamental tool to better support time-dependent and data-intensive applications. In this context, this work explores existing and emerging platforms for MEC and human-centric applications, and proposes a suitable architecture that can be used in the context of autonomous vehicle systems.The proposed architecture will support scalable communication among sensing devices and edge/cloud computing platforms, as well as orchestrate services for computing, storage, and learning with the use of an Information-centric paradigm such as Apache KafkaSource: FRAME '22 - 2nd Workshop on Flexible Resource and Application Management on the Edge, pp. 33–36, Minneapolis, Minnesota, USA, 27/06-01/07/2022
DOI: 10.1145/3526059.3533621
Project(s): TEACHING
Metrics:

See at: ZENODO | dl.acm.org | doi.org | CNR ExploRA

2022 Conference article Open Access

Network measurements with function-as-a-service for distributed low-latency edge applications
Carlini E., Kavalionak H., Dazzi P., Ferrucci L., Coppola M., Mordacchini M.
Edge computing promises to bring computation and storage close to end-users, opening exciting new areas of improvement for applications with a high level of interactivity and requiring low latency. However, these improvements require careful scheduling of applications in the correct Edge resource. This decision is generally taken by considering multiple parameters, including the network capabilities. In this paper, we discuss an approach that measures latency and bandwidth between multiple clients and Edge servers. The approach is based on recent Serverless computing technologies, and it is meant as a support to take timely and correct scheduling decisions in the Edge. We also provide the description of a proof of concept implementation of the said approach.Source: FRAME 2022 - 2nd Workshop on Flexible Resource and Application Management on the Edge (colocated with HPDC 2022), pp. 25–28, Minneapolis, Minnestota, USA, 01/07/2022
DOI: 10.1145/3526059.3533622
Project(s): ACCORDION
Metrics:

See at: ISTI Repository | dl.acm.org | doi.org | CNR ExploRA

2022 Conference article Open Access

See at: ISTI Repository | ceur-ws.org | CNR ExploRA

2022 Conference article Open Access

Decentralized federated learning and network topologies: an empirical study on convergence
Kavalionak H., Carlini E., Dazzi P., Ferrucci L., Mordacchini M., Coppola M.
Federated Learning is a well-known learning paradigm that allows the distributed training of machine learning models. Federated Learning keeps data in the source devices and communicates only the model's coefficients to a centralized server. This paper studies the decentralized flavor of Federated Learning. A peer-to-peer network replaces the centralized server, and nodes exchange model's coefficients directly. In particular, we look for empirical evidence on the effect of different network topologies and communication parameters on the convergence in the training of distributed models. Our observations suggest that small-world networks converge faster for small amounts of nodes, while xx are more suitable for larger setups.Source: SEBD 2022 - 30th Italian Symposium on Advanced Database Systems, pp. 317–324, Tirrenia, Pisa, Italy, 19-22/06/2022
Project(s): TEACHING

See at: ceur-ws.org | ISTI Repository | CNR ExploRA

Driving AI-IoT design towards the UN Sustainable Development Goals (SDGs)
Tulone D., Samuel G., Tibuzzi A., Coppola M., Charter M., Gemma P., Henz P., Gonzales R., Chessa S., Catarci T.
Final technical report from the ITU-T Focus Group on Environmental Efficiency for Artificial Intelligence and other Emerging Technologies (FG-AI4EE), discussing the need for integrating and harmonizing environmental, social models and sustainability needs when designing AI-IoT based solutions. The report highlights (I) current barriers hampering the adoption of a comprehensive path that addresses all three needs, (II) the risks stemming from single-path sustainability approaches, and (III) provides suggestions for future work that can foster and promote the adoption of a more comprehensive process of designing sustainable AI-IoT systems.Source: pp.1–20, 2022
Project(s): TEACHING , ACCORDION

See at: www.itu.int | CNR ExploRA

2021 Conference article Open Access

Inter-operability and orchestration in heterogeneous cloud/edge resources: the ACCORDION vision
Korontanis I., Tserpes K., Pateraki M., Blasi L., Violos J., Ferran D., Marin E., Kourtellis N., Coppola M., Carlini E., Ledwo? Z., Tarkowski P., Loven T., González Rozas Y., Kentros M., Dodis M., Dazzi P.
This paper introduces the ACCORDION framework, a novel frame- work for the management of the cloud-edge continuum, targeting the support of NextGen applications with strong QoE requirements. The framework addresses the need for an ever expanding and het- erogeneous pool of edge resources in order to deliver the promise of ubiquitous computing to the NextGen application clients. This endeavor entails two main technical challenges. First, to assure interoperability when incorporating heterogeneous infrastructures in the pool. Second, the management of the largely dynamic pool of edge nodes. The optimization of the delivered QoE stands as the core driver to this work, therefore its monitoring and modelling comprises a core part of the conducted work. The paper discusses the main pillars that support the ACCORDION vision, and provide a description of the three planned use case that are planned to demonstrate ACCORDION capabilities.Source: FRAME'21 - 1st Workshop on Flexible Resource and Application Management on the Edge, pp. 9–14, Virtual Event, Sweden, 25/06/2021
DOI: 10.1145/3452369.3463816
Project(s): ACCORDION , ACCORDION
Metrics:

See at: ISTI Repository | ZENODO | dl.acm.org | CNR ExploRA

2021 Conference article Open Access

Latency preserving self-optimizing placement at the edge
Ferrucci L., Mordacchini M., Coppola M., Carlini E., Kavalionak H., Dazzi P.
The Internet is experiencing a fast expansion at its edges. The wide availability of heterogeneous resources at the Edge is pivotal in the definition and extension of traditional Cloud solutions toward supporting the development of new applications. However, the dynamic and distributed nature of these resources poses new challenges for the optimization of the behaviour of the system. New decentralized and self-organizing methods are needed to face the needs of the Edge/Cloud scenario and to optimize the exploitation of Edge resources. In this paper we propose a distributed and adaptive solution that reduces the number of replicas of application services that are executed throughout the system, all the while ensuring that the latency constraints of applications are met, thus allowing to also meet the end users' QoS requirements. Experimental evaluations through simulation show the effectiveness of the proposed approach.Source: FRAME'21 - 1st Workshop on Flexible Resource and Application Management on the Edge, pp. 3–8, Virtual Event, Sweden, 25/06/2021
DOI: 10.1145/3452369.3463815
Project(s): ACCORDION , ACCORDION
Metrics:

See at: ISTI Repository | ZENODO | dl.acm.org | CNR ExploRA

2021 Conference article Open Access

See at: ISTI Repository | dl.acm.org | CNR ExploRA

2021 Report Open Access

ACCORDION D4.1 - Edge/Cloud continuum management framework report (I)
Taleb T., Violos J., Tsanakas S., Pagoulatou T., Theodoropoulos T., Coppola M., Dazzi P., Ferrucci L., Diego F., Marin E., Kourtelis N.
This deliverable provides the first report summarizing the scientific advancements, during the first year of the project, achieved by WP4 Tasks. Work Package (WP) 4, dubbed Edge/Cloud continuum management framework, is organized around 6 Tasks is to develop a framework that efficiently manages the deployment and runtime of ACCORDION applications on the continuum.Source: ISTI Project report, ACCORDION, D4.1, 2021
Project(s): ACCORDION

See at: ISTI Repository | CNR ExploRA

2021 Report Open Access

ACCORDION D7.2 - Dissemination & exploitation activities (I)
Tserpes K., Di Girolamo M., Violos J., Lipa B., Paterkai M., Loven T., Tarkowski P., Taleb T., Nadir Z., Kourtelis N., Schmidt S., Ferran D., Dazzi P., Rapisarda B., Coppola M., Vakalellis M.
ACCORDION project aims at unlocking the full potential of a big class of applications that are too latency- sensitive, or data-dependent, to be moved to the public cloud. ACCORDION couples efficient, decentralized and AI-based solutions for cloud and edge resource federation with novel approaches for application definition management and generation at runtime. This deliverable provides a broad overview of the communication, dissemination and exploitation activities implemented within the ACCORDION project within the first 18 months of the project and a detailed market analysis. A detailed exploitation strategy for the project, as well as exploitation plans for individual partners, was presented in the previous deliverable D7.1. Based on those plans, in this deliverable we carry out an evaluation of the ACCORDION project's dissemination and exploitation activities made so far.Source: ISTI Project report, ACCORDION, D7.2, 2021
Project(s): ACCORDION

See at: ISTI Repository | CNR ExploRA

2021 Conference article Open Access

TEACHING - Trustworthy autonomous cyber-physical applications through human-centred intelligence
Bacciu D., Akarmazyan S., Armengaud E., Bacco M., Bravos G., Calandra C., Carlini E., Carta A., Cassarà P., Coppola M., Davalas C., Dazzi P., Degennaro M. C., Di Sarli D., Dobaj J., Gallicchio C., Girbal S., Gotta A., Groppo R., Lomonaco V., Macher G., Mazzei D., Mencagli G., Michail D., Micheli A., Peroglio R., Petroni S., Potenza R., Pourdanesh F., Sardianos C., Tserpes K., Tagliabò F., Vatl J., Varlamis I., Veledar O.
This paper discusses the perspective of the H2020 TEACHING project on the next generation of autonomous applications running in a distributed and highly heterogeneous environment comprising both virtual and physical resources spanning the edge-cloud continuum. TEACHING puts forward a human-centred vision leveraging the physiological, emotional, and cognitive state of the users as a driver for the adaptation and optimization of the autonomous applications. It does so by building a distributed, embedded and federated learning system complemented by methods and tools to enforce its dependability, security and privacy preservation. The paper discusses the main concepts of the TEACHING approach and singles out the main AI-related research challenges associated with it. Further, we provide a discussion of the design choices for the TEACHING system to tackle the aforementioned challenges.Source: IEEE COINS 2021 - IEEE International Conference on Omni-layer Intelligent systems, Online conference, 23-26/08/2021
DOI: 10.1109/coins51742.2021.9524099
DOI: 10.5281/zenodo.5293769
DOI: 10.5281/zenodo.5293768
Project(s): TEACHING
Metrics:

See at: arXiv.org e-Print Archive | ieeexplore.ieee.org | xplorestaging.ieee.org | CNR ExploRA

2021 Conference article Open Access

Impact of network topology on the convergence of decentralized federated learning systems
Kavalionak H., Carlini E., Dazzi P., Ferrucci L., Mordacchini M., Coppola M.
Federated learning is a popular framework that enables harvesting edge resources' computational power to train a machine learning model distributively. However, it is not always feasible or profitable to have a centralized server that controls and synchronizes the training process. In this paper, we consider the problem of training a machine learning model over a network of nodes in a fully decentralized fashion. In particular, we look for empirical evidence on how sensitive is the training process for various network characteristics and communication parameters. We present the outcome of several simulations conducted with different network topologies, datasets, and machine learning models.Source: ISCC 2021 - 26th IEEE Symposium on Computers and Communications, Athens, Greece, 05-08/09/2021
DOI: 10.1109/iscc53001.2021.9631460
Project(s): TEACHING
Metrics:

See at: ISTI Repository | ieeexplore.ieee.org | CNR ExploRA

2021 Conference article Open Access

Self- organizing energy-minimization placement of QoE-constrained services at the edge
Mordacchini M., Ferrucci L., Carlini E., Kavalionak H., Coppola M., Dazzi P.
The wide availability of heterogeneous resources at the Edgeof the network is gaining a central role in defining and developing newcomputing paradigms for both the infrastructures and the applications.However, it becomes challenging to optimize the system's behaviour, dueto the Edge's highly distributed and dynamic nature. Recent solutionspropose new decentralized, self-adaptive approaches to face the needs ofthis scenario. One of the most challenging aspect is related to the opti-mization of the system's energy consumption. In this paper, we proposea fully decentralized solution that limits the energy consumed by thesystem, without failing to match the users expectations, defined as theservices' Quality of Experience (QoE.). Specifically, we propose a schemewhere the autonomous coordination of entities at Edge is able to reducethe energy consumption by reducing the number of instances of the ap-plications executed in system. This result is achieve without violatingthe services' QoE, expressed in terms of latency. Experimental evalua-tions through simulation conducted with PureEdgeSim demonstrate theeffectiveness of the approachSource: GECON 2021: 18th International Conference on Economics of Grids, Clouds, Systems and Services, pp. 133–142, Virtual Event, Rome, 21-23/09/2021
DOI: 10.1007/978-3-030-92916-9_11
Project(s): ACCORDION
Metrics:

See at: ISTI Repository | link.springer.com | CNR ExploRA

2021 Report Open Access

TEACHING D2.2 - Refined requirement specifications and preliminary release of the computing and communication platform
Coppola M., Cassarà P., Gotta A., Dazzi P., Carlini E., Mencagli G., Tserpes K., Girbal S., Valtl J.
This deliverable is aimed at providing an updated report on state-of-the-art analysis, conceptual design and implementation of the Distributed Computing and Communication platform for CPSoS, that we refer as High-Performance Computing and Communication Infrastructure (HPC2I). After providing an update on the State-of-the-art for some of the related research activities we summarize the overall design and features of the HPC2I. We present a more concrete and detailed design of the conceptual, high-level system architecture previously described at M12. Said infrastructure lays the foundation of the TEACHING platform and supports the execution of project use cases. This deliverable also focuses on the implementation of the different software tools and prototypes that compose the HPC2I, highlighting the relationships they have with the work package activities (tasks). For each of these tools is also provided the rationale and a description of the base technologies on top of which the actual software components are developed.Source: ISTI Project report, TEACHING, D2.2, 2021
Project(s): TEACHING

See at: ISTI Repository | CNR ExploRA

2021 Conference article Open Access

Cloud for holography and augmented reality
Makris A., Boudi A., Coppola M., Cordeiro L., Corsini M., Dazzi P., Andilla F. D., Gonzalez Rozas Y., Kamarianakis M., Pateraki M., Pham T. L., Protopsaltis A., Raman A., Romussi A., Rosa L., Spatafora E., Taleb T., Theodoropoulos T., Tserpes K., Zschau E., Herzog U.
The paper introduces the CHARITY framework, a novel framework which aspires to leverage the benefits of intelligent, network continuum autonomous orchestration of cloud, edge, and network resources, to create a symbiotic relationship between low and high latency infrastructures. These infrastructures will facilitate the needs of emerging applications such as holographic events, virtual reality training, and mixed reality entertainment. The framework relies on different enablers and technologies related to cloud and edge for offering a suitable environment in order to deliver the promise of ubiquitous computing to the NextGen application clients. The paper discusses the main pillars that support the CHARITY vision, and provide a description of the planned use cases that are planned to demonstrate CHARITY capabilities.Source: CloudNet 2021 - IEEE 10th International Conference on Cloud Networking, pp. 118–126, Online event, 8-10/11/2021
DOI: 10.1109/cloudnet53349.2021.9657125
Project(s): CHARITY
Metrics:

See at: ZENODO | doi.org | ieeexplore.ieee.org | CNR ExploRA

See at: Future Generation Computer Systems | www.sciencedirect.com | CNR ExploRA