brajesh umrao - Academia.edu (original) (raw)

Papers by brajesh umrao

Research paper thumbnail of Machine Learning for IT Security

Encyclopedia of Machine Learning, 2011

Research paper thumbnail of Placement of virtual network functions for network services

International Journal of Network Management

SummaryNetwork function virtualization (NFV) is a name of technology for replacing hardware‐based... more SummaryNetwork function virtualization (NFV) is a name of technology for replacing hardware‐based network functions with software programs. Virtual network function (VNF) is a software program that replaces the hardware‐based network functionality. The replacement of the hardware‐based network functions (middleboxes) with software programs promises the on‐demand provisioning of network functions and reduces capital and operational expenses of the network. Due to this replaced network can adapt to the different network functions. Network service providers deploy various network services with different objectives, such as reducing the network's active servers and traffic latency or network operational expenses. In this article, a VNF placement problem is studied to optimize the total operating costs of the networks. To solve the VNF placement problem, we proposed an integer linear program (ILP) model, which has been implemented using CPLEX. Although an ILP‐based approach gives an ...

Research paper thumbnail of APVNFC: Adaptive Placement of Virtual Network Function Chains

Cybernetics and Information Technologies

Designing efficient and flexible approaches for placement of Virtual Network Function (VNF) chain... more Designing efficient and flexible approaches for placement of Virtual Network Function (VNF) chains is the main success of Network Function Virtualization (NFV). However, most current work considers the constant bandwidth and flow processing requirements while deploying the VNFs in the network. The constant (immutable) flow processing and bandwidth requirements become critical limitations in an NFV-enabled network with highly dynamic traffic flow. Therefore, bandwidth requirements and available resources of the Point-of-Presence (PoP) in the network change constantly. We present an adaptive model for placing VNF chains to overcome this limitation. At the same time, the proposed model minimizes the number of changes (i.e., re-allocation of VNFs) in the network. The experimental evaluation shows that the adaptive model can deliver stable network services. Moreover, it reduces the significant number of changes in the network and ensures flow performance.

Research paper thumbnail of Algorithms for functionalities of virtual network: a survey

The Journal of Supercomputing, 2021

Nowadays, network functionalities are provided through dedicated hardware middleboxes. These hard... more Nowadays, network functionalities are provided through dedicated hardware middleboxes. These hardware middleboxes are statically embedded with network topology to provide the network services. The types of middleboxes increase in the network, with an increased number of users and network services. The growth in multiple kinds of middleboxes creates an ossified network that is difficult for network management. Network Function Virtualization (NFV) deals with such problems by eliminating network functions from the piece of hardware and implement it as a software to execute in a commodity server. The network virtualization provides more flexibility and elasticity in the network to automatically manage and control the network functions. NFV internally implements the different algorithms to automate the on-demand service provisioning mechanism. In the service provisioning process, NFV allows to share the physical resources among the Virtual Network Functions (VNFs). In order to optimally...

Research paper thumbnail of Comparison of NBTree and VFI Machine Learning Algorithms for Network Intrusion Detection using Feature Selection

International Journal of Computer Applications, 2014

The security of computer networks is of great importance. But, with the proliferation of electron... more The security of computer networks is of great importance. But, with the proliferation of electronic devices and the internet, there has been an exponential rise in malicious activities. The security perpetrators take the advantage of the intricacy of the internet and carry out intrusions. There have been certain researches to find out solutions for detecting intrusions. In this paper, the research has been the application of machine learning techniques to the field of network intrusion detection. Machine learning techniques can learn normal and anomalous patterns from training data and generate classifiers which can be used to detect intrusions in a network. The machine learning techniques used are Naïve Bayes Tree algorithm and the Voting Feature Intervals algorithm. Also, Feature Selection Methods to improve the performance of these algorithms were used because the input to classifiers is in a high dimension feature space, but all features available are not relevant for classification. Two approaches were taken into consideration for feature selection, Chi Square and Gain Ratio. Using these feature selection approaches a comparative study of the two algorithms NBTree and VFI as classifiers has been done. The NSL-KDD data set has been used to train and test the classifiers.

Research paper thumbnail of A survey on optimal utilization of preemptible VM instances in cloud computing

The Journal of Supercomputing

Research paper thumbnail of A comment on sorting by translocations

2013 4th International Conference on Computer and Communication Technology (ICCCT), 2013

ABSTRACT Genome rearrangement is an important research area that studies the ordering of genes in... more ABSTRACT Genome rearrangement is an important research area that studies the ordering of genes in the evolution of gene families. With the advent of large-scale DNA computing genome rearrangements are becoming increasingly important day by day in evolutionary molecular biology. Translocation is one of the important rearrangement event in the evolution of mammalian species as multi chromosomal genome frequently evolves by rearrangement events called translocation. Translocation exchanges genetic material between different chromosomes. Given two multichromosomal genomes A and B, the problem of sorting by translocation is to find a shortest sequence of translocations to transform A into B. It is observed that the distance formula (minimum number of translocations to be used) between genome A and B given by Anne Bergeron et al. fails in some cases. In this paper, we present some of such cases and also provide a modified distance formula.

Research paper thumbnail of Machine Learning for IT Security

Encyclopedia of Machine Learning, 2011

Research paper thumbnail of Placement of virtual network functions for network services

International Journal of Network Management

SummaryNetwork function virtualization (NFV) is a name of technology for replacing hardware‐based... more SummaryNetwork function virtualization (NFV) is a name of technology for replacing hardware‐based network functions with software programs. Virtual network function (VNF) is a software program that replaces the hardware‐based network functionality. The replacement of the hardware‐based network functions (middleboxes) with software programs promises the on‐demand provisioning of network functions and reduces capital and operational expenses of the network. Due to this replaced network can adapt to the different network functions. Network service providers deploy various network services with different objectives, such as reducing the network's active servers and traffic latency or network operational expenses. In this article, a VNF placement problem is studied to optimize the total operating costs of the networks. To solve the VNF placement problem, we proposed an integer linear program (ILP) model, which has been implemented using CPLEX. Although an ILP‐based approach gives an ...

Research paper thumbnail of APVNFC: Adaptive Placement of Virtual Network Function Chains

Cybernetics and Information Technologies

Designing efficient and flexible approaches for placement of Virtual Network Function (VNF) chain... more Designing efficient and flexible approaches for placement of Virtual Network Function (VNF) chains is the main success of Network Function Virtualization (NFV). However, most current work considers the constant bandwidth and flow processing requirements while deploying the VNFs in the network. The constant (immutable) flow processing and bandwidth requirements become critical limitations in an NFV-enabled network with highly dynamic traffic flow. Therefore, bandwidth requirements and available resources of the Point-of-Presence (PoP) in the network change constantly. We present an adaptive model for placing VNF chains to overcome this limitation. At the same time, the proposed model minimizes the number of changes (i.e., re-allocation of VNFs) in the network. The experimental evaluation shows that the adaptive model can deliver stable network services. Moreover, it reduces the significant number of changes in the network and ensures flow performance.

Research paper thumbnail of Algorithms for functionalities of virtual network: a survey

The Journal of Supercomputing, 2021

Nowadays, network functionalities are provided through dedicated hardware middleboxes. These hard... more Nowadays, network functionalities are provided through dedicated hardware middleboxes. These hardware middleboxes are statically embedded with network topology to provide the network services. The types of middleboxes increase in the network, with an increased number of users and network services. The growth in multiple kinds of middleboxes creates an ossified network that is difficult for network management. Network Function Virtualization (NFV) deals with such problems by eliminating network functions from the piece of hardware and implement it as a software to execute in a commodity server. The network virtualization provides more flexibility and elasticity in the network to automatically manage and control the network functions. NFV internally implements the different algorithms to automate the on-demand service provisioning mechanism. In the service provisioning process, NFV allows to share the physical resources among the Virtual Network Functions (VNFs). In order to optimally...

Research paper thumbnail of Comparison of NBTree and VFI Machine Learning Algorithms for Network Intrusion Detection using Feature Selection

International Journal of Computer Applications, 2014

The security of computer networks is of great importance. But, with the proliferation of electron... more The security of computer networks is of great importance. But, with the proliferation of electronic devices and the internet, there has been an exponential rise in malicious activities. The security perpetrators take the advantage of the intricacy of the internet and carry out intrusions. There have been certain researches to find out solutions for detecting intrusions. In this paper, the research has been the application of machine learning techniques to the field of network intrusion detection. Machine learning techniques can learn normal and anomalous patterns from training data and generate classifiers which can be used to detect intrusions in a network. The machine learning techniques used are Naïve Bayes Tree algorithm and the Voting Feature Intervals algorithm. Also, Feature Selection Methods to improve the performance of these algorithms were used because the input to classifiers is in a high dimension feature space, but all features available are not relevant for classification. Two approaches were taken into consideration for feature selection, Chi Square and Gain Ratio. Using these feature selection approaches a comparative study of the two algorithms NBTree and VFI as classifiers has been done. The NSL-KDD data set has been used to train and test the classifiers.

Research paper thumbnail of A survey on optimal utilization of preemptible VM instances in cloud computing

The Journal of Supercomputing

Research paper thumbnail of A comment on sorting by translocations

2013 4th International Conference on Computer and Communication Technology (ICCCT), 2013

ABSTRACT Genome rearrangement is an important research area that studies the ordering of genes in... more ABSTRACT Genome rearrangement is an important research area that studies the ordering of genes in the evolution of gene families. With the advent of large-scale DNA computing genome rearrangements are becoming increasingly important day by day in evolutionary molecular biology. Translocation is one of the important rearrangement event in the evolution of mammalian species as multi chromosomal genome frequently evolves by rearrangement events called translocation. Translocation exchanges genetic material between different chromosomes. Given two multichromosomal genomes A and B, the problem of sorting by translocation is to find a shortest sequence of translocations to transform A into B. It is observed that the distance formula (minimum number of translocations to be used) between genome A and B given by Anne Bergeron et al. fails in some cases. In this paper, we present some of such cases and also provide a modified distance formula.