2690-6791ISSN (Online)
2769-786XISSN (Print)
This paper reports on an in-depth examination of the impact of the backing filesystems to Docker performance in the context of Linux container-based virtualization. The experimental design was a 3x3x4 arrangement, i.e., we considered three different numbers of Docker containers, three filesystems (Ext4, XFS and Btrfs), and four application workloads related to Web server I/O activity, e-mail server I/O activity, file server I/O activity and random file access I/O activity, respectively. The experimental results indicate that Ext4 is the most optimal filesystem, among the considered filesystems, for the considered experimental settings. In addition, the XFS filesystem is not suitable for workloads that are dominated by synchronous random write components (e.g., characteristical for mail workload), while the Btrfs filesystem is not suitable for workloads dominated by random write and sequential write components (e.g., file server workload).
Read MoreDoi: https://doi.org/10.54216/JISIoT.030101
Vol. 3 Issue. 1 PP. 8-17, (2021)
A fall of an older adult often leads to severe injuries and is found to be a significant reason for the death due to post-traumatic complications. Many falls happen in the home atmosphere and prevail unrecognized. Thus, the need for reliable early fall detection is necessary for fast help. Lately, the emergence of wearables, smartphones, IoT, etc., made it possible to develop systems fall detection which aids in the remote monitoring of the elderly. The goal is to allow intelligent algorithms and smartphones to detect falls for elderly care and to monitor them regularly. This work presents the Artificial Intelligence of Things for Fall Detection (AIOTFD) system using a slime mould algorithm (SMA) to optimize the final data. The features extracted using SqueezeNet further CNN based SMA used for data optimization. The validation of the AIOTFD model performance is evaluated through the Multiple Cameras Fall Dataset (MCFD) and UR Fall Detection dataset (URFD). The empirical results accentuated the assuring realization of the model compared to other state-of the art methods.The obtained results shows our proposed AIOTFD attains accuracy of 99.82% and 99.79% and databases can be used for additional investigation and optimizations to increase the recognition rate to enhance the independent life of the elderly.
Read MoreDoi: https://doi.org/10.54216/JISIoT.030102
Vol. 3 Issue. 1 PP. 18-31, (2021)
In the 21st century, the Smart Grid (SG), also known as the next-generation power grid, arose as a substitute for inefficient power systems, ensuring a reliable and efficient power supply. It is projected to improve the reliability and efficiency of energy distribution while having minimal side effects because it is coupled with modern communication and computation capabilities. The huge infrastructure it possesses, as well as the system's underlying communication network, has resulted in a large number of data that necessitates the use of diverse approaches for proper analysis and decision making. When it comes to analyzing this huge amount of data and generating significant insights from it, big data analytics, machine learning (ML), and deep learning (DL), all play a key role. These insights are useful for anomaly detection, fraud detection, price confirmation, fault detection, monitoring energy consumption, and so on. Hence constant and continuous data analysis is an essential part, of the modern smart grid, for its existence. Inspired by providing a reliable and efficient energy distribution, this paper explores and surveys the smart grid architectural elements, ML and DL based applications, and approaches in the context of SG. In addition in terms of ML and DL based data analytics, this paper highlights the limitations of the current research and, highlights future directions as well.
Read MoreDoi: https://doi.org/10.54216/JISIoT.030103
Vol. 3 Issue. 1 PP. 32-42, (2021)
rapid urbanization and the growing population in smart cities pose significant challenges to the management of urban traffic. In recent years, there has been an increasing interest in developing intelligent traffic management systems that leverage advanced machineries, such as the Internet of Things (IoT), and machine learning (ML), to enhance the efficiency and effectiveness of traffic management in smart cities. This paper proposes an intelligent traffic management (ITM) system for smart cities that integrates various computing paradigms to provide real-time traffic information, optimize traffic flow, and improve road safety. The suggested system utilizes an innovative system for the predicting the traffic flows with the goal of enhancing the current level of traffic management in smart cities. An enhanced convolutional autoencoder network is incorporated into the proposed system as a means of extracting the spatial representations contained in traffic flows. Additionally, by the utilization of a refined gated learning module, it possesses the capability of accurately recording temporal dynamics. Our system is evaluated using real-world traffic data, and the results demonstrate its effectiveness in improving traffic flow and reducing congestion in smart cities. Our system has the potential to significantly enhance the performance of traffic management systems in smart cities, decrease traffic crowding, and progress the safety of roads in smart cities.
Read MoreDoi: https://doi.org/10.54216/JISIoT.030104
Vol. 3 Issue. 1 PP. 43-50, (2021)