The new applications populating the Future Internet will increasingly rely on the exchange of data among groups of devices, dynamically established according to their profile and habits (e.g., common interest in same software updates and services). This will definitely challenge traditional group communication solutions that lack the necessary flexibility in group management and do not support effective control policies on involved endpoints (i.e., authorized senders and intended receivers). Indeed, today, Internet can support the following data delivery schemes: unicast, multicast, broadcast, and anycast, according to the way in which the endpoints of the information exchanges are identified. However, there are several reasons exist discouraging network operators to actually offer all such data delivery schemes to end users. As a result, application developers can rely on unicast communications only, and more complex group-based data dissemination policies are implemented as part of specific applications and services and through additional patches to the basic Internet implementation. And yet, group-based communications are crucial in several Internet of Things (IoT) application scenarios. To address the cited issues, the idea of introducing new disruptive network-layer solutions has emerged from recent literature. Among them, in the first part of this work, the Sociocast1 is presented, which has been theorized as enabler of flexible interactions among groups of devices tied by social relationships in the SDN context. While, in order to overcome the problems arising by the centralized nature of MQTT, broker bridging (which has been is now supported by most MQTT implementations), in the second and last part of this work, an optimization technique is proposed. However, broker bridging does not address the performance issues occurring when publishers and subscribers are connected to different brokers. In this work a technique is investigated exploiting the Social Internet of Things (SIoT) concept to decide the most convenient broker for each publisher/subscriber. A hybrid technique which integrates experimental results obtained in an emulated testbed and analytical derivations will be introduced to evaluate the performance of the proposed approach.
Le nuove applicazioni che popoleranno l’Internet del domani faranno sempre più affidamento sullo scambio di dati tra gruppi di dispositivi, stabiliti dinamicamente in base ai loro profili e alle loro abitudini (es. Un comune interesse per gli stessi aggiornamenti software e/o servizi). Ciò metterà sicuramente alla prova le tradizionali soluzioni di comunicazione di gruppo che mancano della flessibilità necessaria nella gestione dei gruppi e non supportano politiche di controllo efficaci sugli endpoint coinvolti (ad esempio, mittenti autorizzati e possibili destinatari). Oggi, infatti, Internet può supportare i seguenti schemi di trasmissione dei dati: unicast, multicast, broadcast e anycast, a seconda del modo in cui vengono identificati gli endpoint degli scambi di informazioni. Tuttavia, esistono diversi motivi che scoraggiano gli operatori di rete a offrire effettivamente tutti questi schemi di data delivery agli utenti finali. Di conseguenza, gli sviluppatori di applicazioni possono fare affidamento solo sulle comunicazioni unicast e le politiche di diffusione dei dati basate su gruppi vengono implementate come parte di applicazioni e servizi specifici e tramite patch aggiuntive all'implementazione di base di Internet. Eppure, le comunicazioni di tipo group based sono cruciali in diversi scenari di applicazioni dell’Internet of Things (IoT). Per affrontare le questioni citate, dalla letteratura recente è emersa l'idea di introdurre nuove soluzioni a livello di rete. Tra queste, nella prima parte di questo lavoro, viene presentato il Sociocast, che è stato teorizzato come abilitatore di interazioni flessibili tra gruppi di dispositivi legati da relazioni sociali nel contesto SDN. Mentre, al fine di superare i problemi derivanti dalla natura centralizzata di MQTT, il broker bridging (che è stato ora supportato dalla maggior parte delle implementazioni MQTT), nella seconda e ultima parte di questo lavoro, viene proposta una tecnica di ottimizzazione. Tuttavia, il broker bridging non risolve i problemi di prestazioni che si verificano quando gli editori e gli abbonati sono collegati a diversi broker. In questo lavoro viene indagata una tecnica che sfrutta il concetto di Social Internet of Things (SIoT) per decidere il broker più conveniente per ogni editore / abbonato. Sarà introdotta una tecnica ibrida che integra i risultati sperimentali ottenuti in un testbed emulato e derivazioni analitiche per valutare le prestazioni dell'approccio proposto.
Enhancing group communications through the social internet of things / Milotta, Giuseppe Massimiliano. - (2021 Apr 16).
Enhancing group communications through the social internet of things
Milotta, Giuseppe Massimiliano
2021-04-16
Abstract
The new applications populating the Future Internet will increasingly rely on the exchange of data among groups of devices, dynamically established according to their profile and habits (e.g., common interest in same software updates and services). This will definitely challenge traditional group communication solutions that lack the necessary flexibility in group management and do not support effective control policies on involved endpoints (i.e., authorized senders and intended receivers). Indeed, today, Internet can support the following data delivery schemes: unicast, multicast, broadcast, and anycast, according to the way in which the endpoints of the information exchanges are identified. However, there are several reasons exist discouraging network operators to actually offer all such data delivery schemes to end users. As a result, application developers can rely on unicast communications only, and more complex group-based data dissemination policies are implemented as part of specific applications and services and through additional patches to the basic Internet implementation. And yet, group-based communications are crucial in several Internet of Things (IoT) application scenarios. To address the cited issues, the idea of introducing new disruptive network-layer solutions has emerged from recent literature. Among them, in the first part of this work, the Sociocast1 is presented, which has been theorized as enabler of flexible interactions among groups of devices tied by social relationships in the SDN context. While, in order to overcome the problems arising by the centralized nature of MQTT, broker bridging (which has been is now supported by most MQTT implementations), in the second and last part of this work, an optimization technique is proposed. However, broker bridging does not address the performance issues occurring when publishers and subscribers are connected to different brokers. In this work a technique is investigated exploiting the Social Internet of Things (SIoT) concept to decide the most convenient broker for each publisher/subscriber. A hybrid technique which integrates experimental results obtained in an emulated testbed and analytical derivations will be introduced to evaluate the performance of the proposed approach.File | Dimensione | Formato | |
---|---|---|---|
Milotta Giuseppe Massimiliano.pdf
accesso aperto
Tipologia:
Tesi di dottorato
Licenza:
DRM non definito
Dimensione
5.93 MB
Formato
Adobe PDF
|
5.93 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.