دانلود مقاله بهینه سازی ابری مبتنی بر محاسبات نرم نوآورانه برای نسل بعدی IoT
خلاصه
1. معرفی
2. آثار مرتبط
3. روش های چارچوب سیستم برای همکاری جدید لبه-ابر
4. تجزیه و تحلیل عملکرد پلت فرم
5. نتیجه گیری
اعلامیه منافع رقابتی
در دسترس بودن داده ها
منابع
Abstract
1. Introduction
2. Related works
3. System framework methods of new edge–cloud collaboration
4. Platform performance analysis
5. Conclusion
Declaration of Competing Interest
Data availability
References
چکیده
هدف این تحقیق کاهش فشار منابع شبکه بر روی مراکز ابری (CC) و گره‌های لبه، بهبود کیفیت خدمات و بهینه‌سازی عملکرد شبکه است. علاوه بر این، نوعی چارچوب همکاری لبه-ابر مبتنی بر اینترنت اشیا (IoT) را مطالعه و طراحی می کند. ابتدا، ماشین‌های کار کارت رزبری پای (RP) به عنوان گره‌های کاری مورد استفاده قرار می‌گیرند و نوعی چارچوب همکاری لبه-ابر برای محاسبات لبه طراحی شده است. این چارچوب عمدتاً از سه لایه شامل RP لبه (ERP)، نظارت و زمان‌بندی RP (MSRP) و CC تشکیل شده است. در میان سه لایه، ارتباط مشارکتی می تواند بین RP ها و بین RP ها و CC ها تحقق یابد. دوم، نوعی الگوریتم تطبیق لبه-ابر در سناریوی محدودیت تاخیر زمانی پیشنهاد شده است. نتایج تحقیق به‌دست‌آمده توسط تکالیف واقعی نشان می‌دهد که تاخیر زمانی کار در تشخیص چهره در حالت همکاری لبه-ابر کمترین است در بین سه حالت کار، از جمله حالت‌های فقط لبه، فقط CC و حالت‌های همکاری لبه-CC، که به تنها 12 ثانیه می‌رسد. . در مقایسه با CC که به تنهایی اجرا می‌شود، نتایج شناسایی نرخ‌های چارچوب در حالت‌های همکاری لبه-ابر و حالت‌های CC هر دو روان‌تر از آنهایی هستند که فقط در حالت لبه هستند، و تشخیص شی در زمان واقعی می‌تواند محقق شود. مصرف کل انرژی اجرای تخلیه توسط کاربران سیستم به طور مداوم با افزایش تعداد کاربران کاهش می یابد. فرض بر این است که تعداد تجهیزات در سیستم ها 150 قطعه است و نرخ صرفه جویی در انرژی سیستم ها تحت تأثیر فرکانس تولید وظیفه است. فرکانس تولید وظیفه با کاهش متناظر در نرخ صرفه جویی انرژی سیستم ها افزایش می یابد. بر اساس تشخیص شی به عنوان مثال، مصرف انرژی سیستم از 18 وات به 16 وات پس از تخصیص الگوریتم ها کاهش می یابد. چارچوب گنجانده شده نرخ بهره‌مندی منابع را بهبود می‌بخشد و مصرف انرژی سیستم را کاهش می‌دهد. علاوه بر این، منابع نظری و عملی را برای اجرای چارچوب همکاری لبه-ابر فراهم می کند.
Abstract
The research aims to reduce the network resource pressure on cloud centers (CC) and edge nodes, to improve the service quality and to optimize the network performance. In addition, it studies and designs a kind of edge–cloud collaboration framework based on the Internet of Things (IoT). First, raspberry pi (RP) card working machines are utilized as the working nodes, and a kind of edge–cloud collaboration framework is designed for edge computing. The framework consists mainly of three layers, including edge RP (ERP), monitoring & scheduling RP (MSRP), and CC. Among the three layers, collaborative communication can be realized between RPs and between RPs and CCs. Second, a kind of edge–cloud matching algorithm is proposed in the time delay constraint scenario. The research results obtained by actual task assignments demonstrate that the task time delay in face recognition on edge–cloud collaboration mode is the least among the three working modes, including edge only, CC only, and edge–CC collaboration modes, reaching only 12 s. Compared with that of CC running alone, the identification results of the framework rates on edge–cloud collaboration and CC modes are both more fluent than those on edge mode only, and real-time object detection can be realized. The total energy consumption of the unloading execution by system users continuously decreases with the increase in the number of users. It is assumed that the number of pieces of equipment in systems is 150, and the energy-saving rate of systems is affected by the frequency of task generation. The frequency of task generation increases with the corresponding reduction in the energy-saving rate of systems. Based on object detection as an example, the system energy consumption is decreased from 18 W to 16 W after the assignment of algorithms. The included framework improves the resource utility rate and reduces system energy consumption. In addition, it provides theoretical and practical references for the implementation of the edge–cloud collaboration framework.
Introduction
In the context of the rapid development of the Internet of Things (IoT), the basic network framework is faced with huge challenges because of the surge in industrial data [1]. The mass production and use of IoT has brought many security issues. Although several organizations have published guidelines for IoT use security, few IoT providers are able to properly follow these guidelines due to lack of accountability [2]. Traditional centralized cloud computing centers cannot deal with massive digital business. In this context, edge computing demonstrates its significant advantages. For example, it can perform the initial analysis of the input data on edge nodes and upload a few data that it cannot process to the cloud center (CC) for processing, which effectively reduces the storage computing and data transmission costs of edge nodes. In addition, the network resource pressures on edge computing nodes and CC are also reduced correspondingly [3]. CC possesses strong storage and computing capacities. The real-time performance of edge servers is significant, and they respond quickly and are flexible. The combination of edge servers with CC to support the 5th generation (5G) basic network can promote the rapid development of domestic manufacturing and accelerate its digital transformation [4], [5].
The edge–cloud collaboration framework is a hot topic of the current research into the IoT. The advantages of the edge–cloud collaboration framework are more obvious, especially when the actual node tasks are considered [6]. After the reinforcement of the processing capacity of edge nodes, they can deal with the uploading of deep learning and other heavy tasks. In addition, actual application scenarios constrain voltage regulation frequency and chip technology to achieve flexible assignment in industry. At the moment, edge nodes unload intensive computing tasks with the constraints of their own resources [7], [8]. Related literature analyses IoT data by the assignment of different user roles and further proposes a kind of IoT data analysis framework structure, which can make the maximum use of cloud resources and then generate corresponding computing models. On the edge sides, the model structure is adopted in the real-time operation on controllers [9], [10], [11].
Conclusion
With the background of IoT, it is difficult to meet the requirements of the high-quality low time delay by delay-sensitive applications because of the long physical distance between CC and sensors. As a result, the service quality can hardly be guaranteed. In this research, cloud computing is combined with edge computing to study and design a kind of edge–cloud collaboration framework. First, RP card machines are utilized as the working nodes in multitask scenarios, and a kind of edge–cloud collaboration framework is designed for edge computing. The framework consists mainly of ERP, MSRP, and CC layers. Among each of the layers, the collaborative communication between RPs and between RP and CC can be realized by wireless networks. Second, a kind of edge–cloud matching algorithm is proposed in time delay constraint scenarios to achieve the deployment of industrial manufacturing lines by initial edge–cloud collaboration frameworks. In addition, the resource utility rate is enhanced, and the system energy consumption is saved. Due to time limits, there are still many disadvantages in the research, and further studies are necessary. During the framework construction in the research, remote command, only remote data streaming transmission, and file transmission are taken into account. In subsequent research, lightweight node tasks will be virtualized combined with KUbemet and lightweight container technologies (such as Docker).