scholarly journals Applying Catastrophe Theory for Network Anomaly Detection in Cloud Computing Traffic

2019 ◽  
Vol 2019 ◽  
pp. 1-11
Author(s):  
Leila Khatibzadeh ◽  
Zarrintaj Bornaee ◽  
Abbas Ghaemi Bafghi

In spite of the tangible advantages of cloud computing, it is still vulnerable to potential attacks and threats. In light of this, security has turned into one of the main concerns in the adoption of cloud computing. Therefore, an anomaly detection method plays an important role in providing a high protection level for network security. One of the challenges in anomaly detection, which has not been seriously considered in the literature, is applying the dynamic nature of cloud traffic in its prediction while maintaining an acceptable level of accuracy besides reducing the computational cost. On the other hand, to overcome the issue of additional training time, introducing a high-speed algorithm is essential. In this paper, a network traffic anomaly detection model grounded in Catastrophe Theory is proposed. This theory is effective in depicting sudden change processes of the network due to the dynamic nature of the cloud. Exponential Moving Average (EMA) is applied for the state variable in sliding window to better show the dynamicity of cloud network traffic. Entropy is used as one of the control variables in catastrophe theory to analyze the distribution of traffic features. Our work is compared with Wei Xiong et al.’s Catastrophe Theory and achieved a maximum improvement in the percentage of Detection Rate in week 4 Wednesday (7.83%) and a 0.31% reduction in False Positive Rate in week 5 Monday. Additional accuracy parameters are checked and the impact of sliding window size in sensitivity and specificity is considered.

Author(s):  
Shruthi P. ◽  
Nagaraj G. Cholli

Cloud Computing is the environment in which several virtual machines (VM) run concurrently on physical machines. The cloud computing infrastructure hosts multiple cloud service segments that communicate with each other using the interfaces. This creates distributed computing environment. During operation, the software systems accumulate errors or garbage that leads to system failure and other hazardous consequences. This status is called software aging. Software aging happens because of memory fragmentation, resource consumption in large scale and accumulation of numerical error. Software aging degrads the performance that may result in system failure. This happens because of premature resource exhaustion. This issue cannot be determined during software testing phase because of the dynamic nature of operation. The errors that cause software aging are of special types. These errors do not disturb the software functionality but target the response time and its environment. This issue is to be resolved only during run time as it occurs because of the dynamic nature of the problem. To alleviate the impact of software aging, software rejuvenation technique is being used. Rejuvenation process reboots the system or re-initiates the softwares. This avoids faults or failure. Software rejuvenation removes accumulated error conditions, frees up deadlocks and defragments operating system resources like memory. Hence, it avoids future failures of system that may happen due to software aging. As service availability is crucial, software rejuvenation is to be carried out at defined schedules without disrupting the service. The presence of Software rejuvenation techniques can make software systems more trustworthy. Software designers are using this concept to improve the quality and reliability of the software. Software aging and rejuvenation has generated a lot of research interest in recent years. This work reviews some of the research works related to detection of software aging and identifies research gaps.


2020 ◽  
Vol 23 (1) ◽  
pp. 24-29 ◽  
Author(s):  
Marcin Zastempowski ◽  
Andrzej Bochat

AbstractThis paper discusses and calculates the impact of the gyroscopic effect on the increase in the agricultural machine working assemblies’ load on bearings of the chaff cutter type. This effect occurs under natural field operational conditions of this type of machine either during the change in direction of its movement or moving over irregular surface and is caused by a sudden change in the axis direction of quickly rotating masses. In a form of graph, there is a presentation of a mathematical model and exemplary results of simulation calculations for selected parameter values related to the movement and operation of machines with a high-speed drum. Calculations were conducted on the basis of analysis of technical data of working machines of the chaff cutter type. Conducted analysis of the agricultural machine working assemblies’ load on bearings showed that these loads may temporarily increase even by ten times in case of the machine turn and eight times in case of moving over irregular of surface, considerably influencing their lifetime.


Author(s):  
Wei Xiong ◽  
Naixue Xiong ◽  
Laurence T. Yang ◽  
Athanasios V. Vasilakos ◽  
Qian Wang ◽  
...  

2013 ◽  
Vol 135 (4) ◽  
Author(s):  
David Noel ◽  
Mathieu Ritou ◽  
Benoit Furet ◽  
Sebastien Le Loch

Angular contact ball bearings are predominantly used for guiding high speed rotors such as machining spindles. For an accurate modeling, dynamic effects have to be considered, most notably in the bearings model. The paper is based on a dynamic model of angular contact ball bearings. Different kinematic hypotheses are discussed. A new method is proposed for the computation of the stiffness matrix: a complete analytical expression including dynamic effects is presented in order to ensure accuracy at high shaft speed. It is demonstrated that the new method leads to the exact solution, contrary to the previous ones. Besides, the computational cost is similar. The new method is then used to investigate the consequence of the kinematic hypotheses on bearing stiffness values. Last, the relevance of this work is illustrated through the computation of the dynamic behavior of a high speed milling spindle. The impact of this new computation method on the accuracy of a finite element spindle model is quantified.


2020 ◽  
Vol 34 (14) ◽  
pp. 2050149
Author(s):  
Ahmad Zamir Chaudhry ◽  
Guang Pan ◽  
Yao Shi

In this paper, water entry process of air launched AUV is investigated by employing fully coupled finite element method and arbitrary Lagrange–Euler formulation (FEM-ALE) and using penalty coupling technique. Numerical model is established to describe the hydrodynamic characteristics and flow patterns of a high-speed water entry AUV. The effectiveness and accuracy of the numerical simulation are verified quantitatively by the experiments of the earlier study. Selection of suitable advection method and mesh convergence study is carried out during experimental validation process. It is found that appropriate mesh size of impact domain is crucial for numerical simulations and second-order Van Leer advection method is more appropriate for high speed water entry problems. Subsequently, the arbitrary Lagrange–Euler (ALE) algorithm is used to describe the variation laws of the impact load characteristics with water entry velocities, water entry angles and different AUV masses. Dimensionless impact coefficient of AUV at different velocities calculated using ALE method is compared with SPH results. This reveals that ALE method can also simulate the water entry process accurately with less computational cost. This research work can provide beneficial reference information for structure design of AUV and for selection of the water entry parameters.


2013 ◽  
Vol 26 (3) ◽  
pp. 308-317 ◽  
Author(s):  
Dingde Jiang ◽  
Cheng Yao ◽  
Zhengzheng Xu ◽  
Wenda Qin

Author(s):  
Bruno L. Dalmazo ◽  
João P. Vilela ◽  
Marilia Curado

This document provides an at-a-glance view of the main contributions of my Ph.D. work. This work aims at improving security and trustworthiness of cloud computing environments by developing a model for predicting cloud network traffic, an approach for detecting anomalies in cloud network traffic that relies on traffic prediction, as well as a mechanism for aggregating similar alarms from an IDS in the context of the cloud network traffic. All the benefits and drawbacks of the contributions were demonstrated in realistic simulations using data from real network traces. Furthermore, the evaluations were conducted with well-known metrics and the results show that all the proposed mechanisms were able to outperform similar proposals in literature.


Sign in / Sign up

Export Citation Format

Share Document