Distributed and scalable computing framework for improving request processing of wearable IoT assisted medical sensors on pervasive computing system

2020 ◽  
Vol 151 ◽  
pp. 257-265 ◽  
Author(s):  
H. Fouad ◽  
Nourelhoda M. Mahmoud ◽  
Mohammed Sayed El Issawi ◽  
Haytham Al-Feel
Author(s):  
А.Н. ВОЛКОВ

Одним из направлений развития сетей связи 5G и сетей связи 2030 является интегрирование в сеть распределенных вычислительных структур, таких как системы пограничных и туманных вычислений (Fog), которые призваны выполнить децентрализацию вычислительной части сетей. В связи с этим необходимо исследовать и определить принципы предоставления услуг на основе распределенной вычислительной инфраструктуры, в том числе в условиях ограниченности ресурсов отдельно взятых составных частей (Fog-устройства). Предлагается новый фреймворк распределенной динамической вычислительной системы туманных вычислений на основе микросервисного архитектурного подхода к реализации, развертыванию и миграции программного обеспечения предоставляемых услуг. Исследуется типовая архитектура микросервисного подхода и ее имплементация в туманные вычисления, а также рассматриваются два алгоритма: алгоритм K-средних для нахождения центра пользовательской нагрузки и алгоритм роевой оптимизации для определения устройства тумана с необходимыми характеристиками для последующей миграции микросервиса. One of the directions of 5G and 2030 communications networks development is the network-integrated distributed structures, such as edge computing (MEC) and Fog computing, which are designed to decentralize the computing part of networks. In this regard, it is necessary to investigate and determine the principles of providing services based on a distributed computing infrastructure, including in conditions of limited resources of individual components (Fog devices). This article proposes a new framework for a distributed dynamic computing system of fog computing based on a microservice architectural approach to the implementation, deployment, and software migration of the services. The article examines the typical architecture of the microservice approach and its implementation in fog computing, and also investigates two algorithms: K-means for finding the center of user load, swarm optimization (PSO) to determine the fog device with the necessary characteristics for the subsequent migration of the microservice.


Author(s):  
Dr. Subarna Shakya

The diverse user demands in the system supported with the internet of things are often managed efficiently, using the computing system that is pervasive. Pervasive computing system in an integration of heterogeneous distributed network and communication technologies and other referred as the ubiquitous computing. All though it handles the user requirement properly. The ingenuousness in the conveyance of the information, in the standard of handling and extending the heterogeneity assistance for the dispersed clients are still under construction in the as it is very challenging in the pervasive computing system. In order to provide proper and a steadfast communication for the users using an IOT based wearable health care device the paper introduces the new dispersed and elastic computing model (DECM). The developed system utilizes the recurrent-learning for the examining the allocation of resources according to the requirements as well as the allotment aspects. Based on the determined requirements of the resources, the pervasive computing system provide services to the user in the end with minimized delay and enhanced rate of communication for the health care wearable devices. The developed system emphasis also on managing the mobility, apart from allocation of resources and distribution for proper data conveyance over the wearable health care device. The working of the laid out system is determined by the experimental analysis. The constancy of the model developed is demonstrated utilizing the metrics such as the failure of request, time of response, managed and backlogged requests, bandwidth as well as storage used. The developed model heightens the number of request managed properly (handled) along with the bandwidth and storage and minimizes the failure in requests, backlogs and the time taken for response.


To build the productivity of every errand, we necessitate a framework that should furnish high performance alongside adaptabilities and price effectiveness for client. Cloud computing, since we are for the most part mindful, has turned out to be well known over the previous decade. So as to build up a high performance disseminated framework, we have to use the cloud computing. In this paper, we will initially have a presentation of high performance computing framework. Thusly inspecting them we will investigate inclines in compute and emerald feasible computing to upgrade the routine of a cloud framework. At long last introducing the future degree, we finish up the paper recommending a way to accomplish a emerald high performance cloud framework.


Sign in / Sign up

Export Citation Format

Share Document