Search

A cloud is made up of many data centers, with its own set of data and resources. The reasons for employing several cloud sites to operate a workflow are that the data is already dispersed, the required resources surpass the constraints of a single site. This paper presents a hybrid multi-objective optimization algorithm denoted as HSOS-SOA, achieved by combining the Symbiotic Organisms Search and Seagull Optimization Algorithm. The HSOS-SOA uses chaotic maps to generate random numbers and performs a good trade-off between exploration and exploitation, resulting in a higher convergence rate. HSOS-SOA is used to solve scientific workflow scheduling problems in multisite cloud computing by taking into consideration elements such as makespan, cost, and reliability. A solution is chosen from the Pareto front using the knee-point approach in this approach. Extensive analyses are performed out in Microsoft Azure multisite cloud and the results exhibited that the HSOS-SOA can outperform other algorithms in terms of metrics such as IGD, Coverage Ratio, and so on. Experimental results of experiments reveal that the results in makespan improvement in the range of 5.72–28.61%, cost in the range of 5.16–45.16%, and reliability in the range of 3.11–25% over well-known metaheuristic algorithms. [ABSTRACT FROM AUTHOR]

Traffic prediction is critical to expanding a smart city and country because it improves urban planning and traffic management. This prediction is very challenging due to the multifactorial and random nature of traffic. This study presented a method based on ensemble learning to predict urban traffic congestion based on weather criteria. We used the NAS algorithm, which in the output based on heuristic methods creates an optimal model concerning input data. We had 400 data, which included the characteristics of the day's weather, including six features: absolute humidity, dew point, visibility, wind speed, cloud height, and temperature, which in the final column is the urban traffic congestion target. We have analyzed linear regression with the results obtained in the project; this method was more efficient than other regression models. This method had an error of 0.00002 in terms of MSE criteria and SVR, random forest, and MLP methods, which have error values of 0.01033, 0.00003, and 0.0011, respectively. According to the MAE criterion, this method has a value of 0.0039. The other methods have obtained values of 0.0850, 0.0045, and 0.027, respectively, which show that our proposed model has a minor error than other methods and has been able to outpace the other models. [ABSTRACT FROM AUTHOR]

One of the technology for increasing the safety and welfare of humans in roads is Vehicular Cloud Computing (VCC). This technology can utilize cloud computing advantages in the Vehicular Ad Hoc Network (VANET). VCC by utilizing modern equipment along with current vehicles, can play a significant role in the area of smart transportation systems. Given the potential of this technology, effective methods for managing existing resources and providing the expected service quality that is essential for such an environment are not yet available as it should. One of the most important barriers to providing such solutions seems to be resource constraints and very high dynamics in vehicles in VCC. In this article, based on virtualization and taking into account the environment with these features, we propose simple ways to manage resources better and improve the quality of service. We were able to achieve better results in simulation than previous methods by providing a flexible data structure to control the important data in the environment effectively. To illustrate the impact of the proposed methods, we compared them with some of the most important methods in this context, and we used SUMO 1.2.0 and MATLAB R2019a software to simulate them. Simulation results indicate that the proposed methods provide better results than previous methods in terms of resource efficiency, Quality of Service (QoS), and load balancing. [ABSTRACT FROM AUTHOR]

Efficient data management is a key issue for environments distributed on a large scale such as the data cloud. This can be taken into account by replicating the data. The replication of data reduces the time of service and the delay in availability, increases the availability, and optimizes the distribution of load in the system. It is worth mentioning, however, that with the replication of data, the use of resources and energy increases due to the storing of copies of the data. We suggest a replication manager that decreases the cost of using resources, energy, and the delay in the system, and also increases the availability of the system. To reach this aim, the suggested replication manager, called the locality replication manager (LRM), works by using two important algorithms that use the physical adjacency feature of blocks. In addition, a set of simulations are reported to show that LRM can be a suitable option for distributed systems as it uses less energy and resources, optimizes the distribution of load, and has more availability and less delay. [ABSTRACT FROM AUTHOR]