Department of Information Technology, Hazara University Mansehra, Mansehra 21120, Pakistan.
Department of Telecommunication, Hazara University Mansehra, Mansehra 21120, Pakistan.
Sensors (Basel). 2022 Jun 30;22(13):4969. doi: 10.3390/s22134969.
Cloud computing coupled with Internet of Things technology provides a wide range of cloud services such as memory, storage, computational processing, network bandwidth, and database application to the end users on demand over the Internet. More specifically, cloud computing provides efficient services such as "". However, Utility providers in Smart Grid are facing challenges in the design and implementation of such architecture in order to minimize the cost of underlying hardware, software, and network services. In Smart Grid, smart meters generate a large volume of different traffics, due to which efficient utilization of available resources such as buffer, storage, limited processing, and bandwidth is required in a cost-effective manner in the underlying network infrastructure. In such context, this article introduces a QoS-aware Hybrid Queue Scheduling (HQS) model that can be seen over the IoT-based network integrated with cloud environment for different advanced metering infrastructure (AMI) application traffic, which have different QoS levels in the Smart Grid network. The proposed optimization model supports, classifies, and prioritizes the AMI application traffic. The main objective is to reduce the cost of buffer, processing power, and network bandwidth utilized by AMI applications in the cloud environment. For this, we developed a simulation model in the CloudSim simulator that uses a simple mathematical model in order to achieve the objective function. During the simulations, the effects of various numbers of cloudlets on the cost of virtual machine resources such as RAM, CPU processing, and available bandwidth have been investigated in cloud computing. The obtained simulation results exhibited that our proposed model successfully competes with the previous schemes in terms of minimizing the processing, memory, and bandwidth cost by a significant margin. Moreover, the simulation results confirmed that the proposed optimization model behaves as expected and is realistic for AMI application traffic in the Smart Grid network using cloud computing.
云计算与物联网技术相结合，为终端用户提供了广泛的云服务，例如内存、存储、计算处理、网络带宽和数据库应用程序，这些服务都可以通过互联网按需提供。具体来说，云计算提供了高效的服务，例如“”。然而，智能电网中的实用程序提供商在设计和实现这种架构时面临着挑战，以便最大限度地降低底层硬件、软件和网络服务的成本。在智能电网中，智能仪表会生成大量不同的流量，因此需要在底层网络基础设施中以经济有效的方式高效利用可用资源，如缓冲区、存储、有限的处理能力和带宽。在这种情况下，本文介绍了一种 QoS 感知混合队列调度 (HQS) 模型，该模型可以在基于物联网的网络中看到，该网络与云环境集成在一起，用于不同的高级计量基础设施 (AMI) 应用程序流量，这些流量在智能电网网络中具有不同的 QoS 级别。所提出的优化模型支持、分类和优先处理 AMI 应用程序流量。主要目标是降低 AMI 应用程序在云环境中使用的缓冲区、处理能力和网络带宽的成本。为此，我们在 CloudSim 模拟器中开发了一个仿真模型，该模型使用简单的数学模型来实现目标函数。在仿真过程中，研究了在云计算中，各种数量的云服务对虚拟机资源（如 RAM、CPU 处理和可用带宽）成本的影响。仿真结果表明，我们提出的模型在最小化处理、内存和带宽成本方面比以前的方案具有显著的优势。此外，仿真结果证实了所提出的优化模型在使用云计算的智能电网网络中的 AMI 应用程序流量方面表现良好，并且符合实际情况。
