Towards Distributed Mobile Computing

Abstract

Abstract-In the latest years, we observed an exponential growth of the market of the mobile devices. In this scenario, it assumes a particular relevance the rate at which mobile devices are replaced. According to the International Telecommunicaton Union in fact, smart-phone owners replace their device every 20 months, on average. The side effect of this trend is to deal with the disposal of an increasing amount of electronic devices which, in many cases, are still working. We believe that it is feasible to recover such an unexploited computational power. Through a change of paradigm in fact, it is possible to achieve a two-fold objective: 1) extend the mobile devices lifetime; 2) enable a new opportunity to speed up mobile applications. In this paper we aim at providing a survey of state-of-art solutions aim at going in the direction of a Distributed Mobile Computing paradigm. We put in evidence the challenges to be addressed in order to implement this paradigm and we propose some possible future improvements.

BarMan: A run-time management framework in the resource continuum

“…While research concentrated on architectural aspects of Fog systems, not many proposals provide complete and available frameworks able to implement such architectures [12]. In particular, two main aspects need to be considered in evaluating the frameworks: (1) how the application can be developed and integrated, (2) how the resources are shared and allocated to the application. The first aspect is related to the availability of a suitable programming model; the latter requires a system to manage applications and Fog devices.…”

“…All the parameters are then inserted in the objective function (1) where y n is the optimization variable that indicates if the n−DNN is selected. The function considers both DNN mean latency and accuracy to select the best DNNs available depending on the desired behavior expressed by the value of γ.…”

“…. ) can be interconnected to acquire increasing amounts of data that are further processed remotely for a variety of purposes [1]. The numbers are quite impressive: according to IDC [2], there will be 60 billion of connected entities by 2025, generating about 80 ZB of raw data.…”

Back to the future

“…The requirements of the emerging scenarios, together with the limitations of Cloud-based solutions, ask for an intermediate processing layer, that could fill part of the performance gap between Edge/IoT devices and Cloud computing servers. This novel concept, called Fog Computing, was proposed in 2012 by Cisco [6], and it has been translated into different paradigms [24], including Mobile Edge Computing [8], Dew Computing [21] and Mobile Cloud Computing [9,15]. Fog Computing aims at building a scalable horizontal computing infrastructure closer to the IoT layer.…”

Plataforma colaborativa, distribuida, escalable y de bajo costo basada en microservicios, contenedores, dispositivos móviles y servicios en la Nube para tareas de cómputo intensivo

“…Extendiendo el análisis anterior, Pavel Mach y Zdenek Becvar [MAC17], y Giacomo Massari y su equipo [MAS16] estudiaron los casos de aplicación, escenarios y arquitecturas posibles para el aprovechamiento de las capacidades de la computación móvil en el borde. Mariela Curiel y su equipo [CUR18] presentaron una plataforma para el procesamiento paralelo de imágenes médicas, utilizando cómputo móvil sobre una arquitectura GRID, basada en el proyecto BOINC [AND04].…”